doc_id
string | appl_id
string | claim1
string | claim_number
int64 | claim_text
string |
|---|---|---|---|---|
9740696
|
14203229
|
1. <nc>A method</nc> performed by <nc>one or more computers</nc>, <nc>the method</nc> comprising: receiving, by <nc>the one or more computers</nc>, <nc>audio data</nc> from <nc>a client device</nc>; determining, by <nc>the one or more computers</nc>, <nc>specific content</nc> from <nc>captured media</nc> <nc>that</nc> matches <nc>the received audio data</nc>, wherein <nc>the captured media</nc> is captured from <nc>one or more media sources</nc> at <nc>a backend system</nc> and includes at least one of <nc>audio media</nc> or <nc>audio-video media</nc>, and wherein determining <nc>the specific content</nc> <nc>that</nc> matches <nc>the received audio data</nc> includes identifying <nc>an advertisement</nc> <nc>that</nc> is <nc>(i</nc>) included within <nc>the captured media</nc> and <nc>(ii</nc>) included within <nc>the media programming</nc> associated with <nc>the received audio data</nc>, wherein <nc>the captured media</nc> comprises <nc>a collection</nc> of <nc>audio fingerprints</nc> stored in <nc>an audio fingerprint repository</nc>, wherein <nc>each audio fingerprint</nc> is generated from <nc>at least a portion</nc> of <nc>an audio stream</nc> extracted from <nc>one or more monitored digital television broadcast channels</nc>; obtaining, by <nc>the one or more computers</nc>, additional information associated with <nc>the advertisement</nc> included within <nc>the media programming</nc> associated with <nc>the received audio data</nc>; generating, by <nc>the one or more computers</nc>, <nc>a search query</nc> based at least in <nc>part</nc> on <nc>the obtained additional information</nc> associated with <nc>the advertisement</nc> included within <nc>the media programming</nc> associated with <nc>the received audio data</nc>; and returning, by <nc>the one or more computers</nc>, <nc>one or more search results</nc> to <nc>the client device</nc> responsive to <nc>the search query</nc>.
| 10
|
10. <nc>The method</nc> of <nc>claim</nc> 1 , wherein obtaining <nc>the additional information</nc> comprises obtaining <nc>metadata</nc> derived from <nc>the advertisement</nc> included within <nc>the media programming</nc> associated with <nc>the received audio data</nc>; and wherein generating <nc>the search query</nc> comprises generating <nc>the search query</nc> based at least in <nc>part</nc> on <nc>the metadata</nc> derived from <nc>the advertisement</nc> included within <nc>the media programming</nc> associated with <nc>the received audio data</nc>.
|
10038730
|
14725895
|
1. <nc>A non-transitory computer storage medium storing computer-useable instructions</nc> <nc>that</nc>, when used by <nc>one or more computing devices</nc>, cause <nc>the one or more computing devices</nc> to perform <nc>operations</nc> for facilitating <nc>portability</nc> of <nc>web meeting interactions</nc>, <nc>the operations</nc> comprising: receiving, from <nc>a client device</nc> <nc>that</nc> is associated with <nc>a participant identifier</nc>, <nc>at least a first annotating input</nc> for <nc>a first instance</nc> of <nc>a web meeting identifier</nc>, wherein <nc>the received first annotating input</nc> is stored independently from <nc>a recording</nc> of <nc>a first web meeting</nc> <nc>that</nc> corresponds to <nc>the first instance</nc>; processing <nc>the received first annotating input</nc> to include therein <nc>a first unique descriptor</nc> <nc>that</nc> references <nc>the participant identifier</nc>, <nc>the first instance</nc>, <nc>a time</nc>, and <nc>the recording</nc>; generating <nc>an interaction record</nc> <nc>that</nc> includes <nc>the processed first annotating input</nc>; transmitting, to <nc>the client device</nc>, <nc>at least a first portion</nc> of <nc>the interaction record</nc> <nc>that</nc> includes <nc>the processed first annotating input</nc>; receiving, from <nc>the client device</nc> to export to <nc>a second instance</nc> of <nc>the web meeting identifier</nc>, <nc>the processed first annotating input</nc> selected from <nc>the transmitted first portion</nc> of <nc>the interaction record</nc>; generating <nc>a second annotating input</nc> <nc>that</nc> is based on <nc>a portion</nc> of <nc>the selected first annotating input</nc>, <nc>the generated second annotating input</nc> including <nc>a second unique descriptor</nc> <nc>that</nc> at least references <nc>the second instance</nc>; and providing, to <nc>the client device</nc> for <nc>the second instance</nc> of <nc>the web meeting identifier</nc>, <nc>at least a second portion</nc> of <nc>the interaction record</nc> <nc>that</nc> includes <nc>the generated second annotating input</nc>, <nc>the generated second annotating input</nc> to be provided in <nc>a second web meeting</nc> <nc>that</nc>, when instantiated, corresponds to <nc>the second instance</nc>, wherein <nc>the first meeting</nc> and <nc>the second web meeting</nc> are <nc>different occurrences</nc>.
| 2
|
2. <nc>The medium</nc> of <nc>claim</nc> 1 , wherein <nc>the web meeting identifier</nc> is <nc>a topical framework</nc> from <nc>which</nc> <nc>all instances</nc> thereof are based.
|
9536269
|
13349807
|
1. <nc>A customer predictive experience platform</nc>, comprising: <nc>a processor</nc> and <nc>memory</nc>, cooperating to function as: <nc>an outcome engine</nc> configured for <nc>information mining</nc> and for applying <nc>rules</nc> and <nc>analytics</nc> to said <nc>information</nc>; <nc>an ops module</nc> configured for providing <nc>agent performance management</nc>, <nc>average handling time (AHT) analytics</nc>, <nc>workflow management</nc> (<nc>WFM</nc>), and <nc>voice</nc> of the customer (<nc>VoC</nc>) facilities; <nc>a chat module</nc>; <nc>a social media dialog engine</nc>; and <nc>a solution client</nc>, interactive with <nc>said outcome engine</nc>, configured for effecting: <nc>predictive self-service</nc> by automatically serving <nc>curated contents</nc> and <nc>chat</nc> to implement <nc>any</nc> of <nc>social media brand improvement</nc>; <nc>active auto sentiment management</nc> and <nc>rapid response</nc> by identifying and counteracting <nc>negative sentiment</nc>, identifying and reinforcing <nc>positive sentiment</nc>, and <nc>building authority</nc>; <nc>a customer experience ticker</nc>; <nc>a pre- and post-launch pulse</nc>; <nc>enhanced brand ambassadors</nc>; <nc>integration</nc> into <nc>corporate messaging</nc> and <nc>marketing</nc>, including <nc>two-way information exchange</nc> and <nc>synergy</nc>, by leveraging <nc>information flow</nc> across <nc>channels</nc> and providing <nc>a unified view</nc> of <nc>content</nc> and <nc>state</nc>; <nc>a social media dashboard</nc>; and <nc>a live portal</nc> configured for <nc>social media engagement</nc> and <nc>feedback</nc>.
| 2
|
2. <nc>The platform</nc> of <nc>claim</nc> 1 , <nc>the solution client</nc> further configured for: extracting <nc>a real time customer satisfaction value</nc> from <nc>a real-time survey score</nc> from <nc>any</nc> of <nc>customers</nc> receiving <nc>services</nc>, general and/or issue specific; friends activities, updates, and comments; <nc>real time sentiment comparison</nc>; <nc>live feed</nc> with <nc>regard</nc> to <nc>social media tweets</nc>, <nc>marketing</nc>; <nc>live feed</nc> with <nc>regard</nc> to <nc>a customer experience ticker</nc> <nc>that</nc> provides <nc>real feedback</nc> in <nc>real time</nc>; and live <nc>feed</nc> with <nc>regard</nc> to <nc>positive blogs</nc>.
|
8781204
|
12602227
|
1. A method for checking <nc>the authenticity</nc> of <nc>security documents</nc>, in <nc>particular banknotes</nc>, wherein <nc>authentic security documents</nc> comprise <nc>security features</nc> printed, applied or otherwise provided on <nc>the security documents</nc>, <nc>which</nc> <nc>security features</nc> comprise <nc>characteristic visual features</nc> intrinsic to <nc>the processes</nc> used for producing <nc>the security documents</nc>, wherein <nc>the method</nc> comprises <nc>the steps</nc> of: acquiring <nc>a sample image</nc> of <nc>at least one region</nc> of <nc>interest</nc> of <nc>the surface</nc> of <nc>a candidate document</nc> to be authenticated, <nc>which region</nc> of <nc>interest</nc> encompasses <nc>at least part</nc> of <nc>said security features</nc>; <nc>digitally processing</nc> said <nc>sample image</nc> by performing <nc>a decomposition</nc> of <nc>the sample image</nc> into <nc>at least one scale sub</nc><nc>-</nc><nc>space</nc> containing <nc>high resolution details</nc> of <nc>the sample image</nc> and extracting classifying <nc>features</nc> from said <nc>scale</nc> sub<nc>-</nc><nc>space</nc>, <nc>which</nc> extracted classifying <nc>features</nc> are used to position <nc>the candidate document</nc> in <nc>a feature space</nc> enabling <nc>a classification</nc> of <nc>the candidate document</nc>; and deriving <nc>an authenticity rating</nc> of <nc>the candidate document</nc> based on the extracted classifying <nc>features</nc> and <nc>the positioning</nc> of <nc>the candidate document</nc> in <nc>the feature space</nc>.
| 10
|
10. <nc>The method</nc> according to <nc>claim</nc> 1 , wherein said <nc>sample image</nc> is acquired at <nc>a resolution</nc> lower than 600 dpi, preferably of <nc>300 dpi</nc>.
|
9582489
|
14968175
|
1. <nc>A system</nc> for correcting <nc>a phonetically sourced spelling mistake</nc> comprising: <nc>a processor</nc>; and <nc>a memory</nc> coupled to <nc>the processor</nc>, wherein <nc>the memory</nc> comprises <nc>instructions</nc> <nc>which</nc> when executed by <nc>the processor</nc> cause <nc>the processor</nc> to implement <nc>phonetic edit distance context match module</nc> for correcting <nc>a phonetic spelling mistake</nc>, <nc>the phonetic edit distance context match module</nc> comprising: <nc>a language text string buffer</nc> for receiving <nc>a text string</nc> including <nc>a spelling mistake word</nc>; <nc>a phonetic transcription engine</nc> executing on <nc>the processor</nc> for transcribing <nc>the spelling mistake word</nc> into <nc>a phonetic transcription</nc> using <nc>a phonetic dictionary</nc>; <nc>an edit distance engine</nc> executing on <nc>the processor</nc> for determining <nc>a set</nc> of <nc>correctly spelled phonetic forms</nc> from <nc>a phonetic form dictionary</nc> having <nc>shortest edit distances</nc> between <nc>the correctly spelled phonetic forms</nc> and <nc>the phonetic transcription</nc>, wherein <nc>the phonetic form dictionary</nc> comprises correctly spelled <nc>words</nc> and associated phonetic forms; <nc>a context engine</nc> executing on <nc>the processor</nc> for determining <nc>a context probability</nc> of <nc>each correctly spelled phonetic form</nc> being <nc>a correctly spelled word</nc> in <nc>a context</nc> of <nc>other words</nc> in <nc>the text string</nc> using <nc>a statistical language model</nc>; and <nc>a correction engine</nc> executing on <nc>the processor</nc> for substituting <nc>a correctly spelled word</nc> corresponding to <nc>a correctly spelled phonetic form</nc> having <nc>a highest context probability</nc> for <nc>the spelling mistake word</nc> in <nc>the text string</nc>.
| 8
|
8. <nc>A system</nc> according to <nc>claim</nc> 1 , wherein <nc>the spelling mistake word</nc> comprises <nc>any language characters</nc> from <nc>a language</nc> having <nc>a phonetic basis</nc>.
|
8666367
|
12434586
|
1. <nc>A computer-implemented method</nc> performed by <nc>a mobile device</nc>, <nc>the method</nc> comprising: accessing, by <nc>the mobile device</nc>, <nc>a notification service</nc> on <nc>a server</nc> separate from <nc>the mobile device</nc>, <nc>the notification service</nc> hosting <nc>a plurality</nc> of <nc>command collection topics</nc>, where <nc>a distinct mobile device</nc> is subscribed to <nc>each command collection topic</nc>; accessing, by <nc>the mobile device</nc>, <nc>a command collection topic</nc> hosted on <nc>the notification service</nc> and subscribed to by <nc>the mobile device</nc>; <nc>polling</nc>, by <nc>the mobile device</nc>, <nc>the command collection topic</nc> subscribed to by <nc>the mobile device</nc> to determine that <nc>one or more new remote command messages</nc> have been received by <nc>the command collection topic</nc> subscribed to by <nc>the mobile device</nc>; retrieving, by <nc>the mobile device</nc>, in <nc>response</nc> to <nc>the determining</nc> that <nc>one or more new remote command messages</nc> have been received by <nc>the command collection topic</nc>, at least one of <nc>the one or more new remote command messages</nc> included in <nc>the command collection topic</nc> subscribed to by <nc>the mobile device</nc>, wherein <nc>the one or more new remote command messages</nc> identify <nc>commands</nc> to be executed by <nc>the mobile device</nc>; determining, by <nc>the mobile device</nc>, whether <nc>the command</nc> identified by <nc>the retrieved remote command message</nc> can be executed by <nc>the mobile device</nc>; <nc>publishing</nc>, by <nc>the mobile device</nc>, <nc>a result message</nc> associated with <nc>the command</nc> to <nc>a result topic</nc> hosted on <nc>the notification service</nc>; and selectively executing, by <nc>the mobile device</nc>, <nc>the command</nc> based on <nc>a result</nc> of <nc>the determining</nc>.
| 2
|
2. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , wherein publishing <nc>the result message</nc> further comprises: identifying in <nc>the remote command</nc> message <nc>the result topic</nc> hosted on <nc>the notification service</nc>.
|
8925041
|
12968203
|
1. <nc>A computer program product</nc>, comprising <nc>a non-transitory computer usable medium</nc> having <nc>a computer readable program code</nc> embodied therein, <nc>the computer readable program code</nc> adapted to be executed to cause <nc>a computer</nc> to implement <nc>a method</nc>, comprising: identifying <nc>a plurality</nc> of <nc>users</nc> associated with <nc>a client</nc> of <nc>a system</nc>, where <nc>the client</nc> of <nc>the system</nc> uses <nc>a portal</nc> to limit <nc>access</nc> to <nc>client data</nc> within <nc>the system</nc> by <nc>each</nc> of <nc>the plurality</nc> of <nc>users</nc>; identifying <nc>one or more user permissions</nc> for <nc>each</nc> of <nc>the plurality</nc> of <nc>users</nc> <nc>that</nc> enable <nc>access</nc> to <nc>predetermined client data</nc> within <nc>the system</nc> by <nc>each</nc> of <nc>the plurality</nc> of <nc>users</nc> via <nc>the portal</nc>; identifying <nc>one or more sharing rules</nc> for <nc>the client</nc> of <nc>the system</nc> <nc>that</nc> enable <nc>access</nc> to <nc>predetermined client data</nc> within <nc>the system</nc> by <nc>each</nc> of <nc>the plurality</nc> of <nc>users</nc> via <nc>the portal</nc>; and notifying <nc>the client</nc> of <nc>both the user permissions</nc> and <nc>the sharing rules</nc>, in <nc>addition</nc> to <nc>the predetermined client data</nc> to <nc>which</nc> <nc>access</nc> is enabled via <nc>the portal</nc>.
| 6
|
6. <nc>The computer program product</nc> of <nc>claim</nc> 1 , wherein notifying <nc>the client</nc> includes displaying <nc>a report</nc> to <nc>the client</nc>.
|
9660869
|
14533968
|
1. <nc>A method</nc> for <nc>implementation</nc> by <nc>one or more data processors</nc> forming <nc>part</nc> of <nc>at least one computing system</nc>, <nc>the method</nc> comprising: obtaining, by <nc>at least one data processor</nc>, <nc>a plurality</nc> of <nc>records</nc> from <nc>a plurality</nc> of <nc>sources</nc>, <nc>the plurality</nc> of <nc>records</nc> comprising <nc>a plurality</nc> of <nc>types</nc> of <nc>data</nc>; assembling, by <nc>at least one data processor</nc>, <nc>a plurality</nc> of <nc>typed datasets</nc> based on <nc>the obtained records</nc>, <nc>the assembling</nc> comprising: extracting, from <nc>the plurality</nc> of <nc>records</nc>, typed <nc>data</nc> <nc>that</nc> corresponds to <nc>all data</nc> of <nc>a single type</nc> found in <nc>the obtained records</nc>; assembling, by <nc>at least one data processor</nc>, at least one network comprising: <nc>a plurality</nc> of <nc>nodes</nc> representing <nc>all instances</nc> of <nc>the typed data</nc> corresponding to <nc>a common event</nc>; and <nc>a plurality</nc> of <nc>edges</nc> representing <nc>a relationship</nc> between <nc>the plurality</nc> of <nc>nodes</nc>, <nc>the relationship</nc> defining <nc>a connection</nc> between two or more of <nc>the plurality</nc> of <nc>nodes</nc>, where <nc>the edges</nc> comprise <nc>a weighting attribute</nc> representing <nc>a similarity</nc> between <nc>the nodes</nc> connected by <nc>the connection</nc>, <nc>the plurality</nc> of <nc>nodes</nc> and <nc>the plurality</nc> of <nc>edges</nc> stored as <nc>accessible memory objects</nc> in <nc>the at least one computing system</nc>; assembling, by <nc>at least one data processor</nc>, <nc>a vector</nc> by using <nc>a network analyzer</nc>, <nc>the assembling</nc> comprising: determining <nc>a required input format</nc> for <nc>an analytic</nc> configured to operate on <nc>the vector</nc>; and generating, at <nc>the network analyzer</nc>, <nc>the vector</nc> comprising <nc>a subset</nc> of <nc>the typed data</nc> corresponding to <nc>the required input format</nc>; passing, by <nc>at least one data processor</nc>, <nc>the vector</nc> to the analytic; generating, by <nc>at least one data processor</nc> and <nc>the analytic</nc>, <nc>an output</nc> from <nc>the analytic</nc> based on at least <nc>the vector</nc> passed to the analytic, <nc>the output</nc> comprising <nc>electronic data</nc> corresponding to <nc>a result</nc> of <nc>the analytic operating</nc> on <nc>the vector</nc>; and providing, by <nc>at least one data processor</nc>, data comprising <nc>the output</nc>.
| 7
|
7. <nc>The method</nc> of <nc>claim</nc> 1 , wherein the analytic includes <nc>an edge analytic</nc> configured to perform <nc>operations</nc> comprising: receiving <nc>an edge reference</nc> and at least two node references; and generating, from <nc>the edge reference</nc> and <nc>the at least two node references</nc>, <nc>a response</nc> to <nc>a query</nc> corresponding to <nc>a feature</nc> common to <nc>the at least two node references</nc>.
|
10095778
|
14792214
|
1. <nc>A computer-implemented method</nc> of operating <nc>a computerized search engine</nc> to identify and rank <nc>relevant documents</nc> from <nc>a corpus</nc> comprising <nc>multiple millions</nc> of <nc>citationally-related source documents</nc>, said <nc>computer-implemented method</nc> comprising: storing on <nc>a computer-readable storage device</nc> of <nc>the computerized search engine</nc> <nc>a search index</nc> comprising <nc>a first set</nc> of <nc>identification information</nc> identifying <nc>potential input documents</nc> selected from <nc>said source documents</nc> and, for <nc>each</nc> said <nc>potential input document</nc>, <nc>a second set</nc> of <nc>identification information</nc> identifying <nc>a selected number</nc> of <nc>citationally-related potential output documents</nc> selected from <nc>said source documents</nc>; calculating, via <nc>one or more computer-processors</nc> coupled to said <nc>computer-readable storage device</nc>, <nc>a first numerical score</nc> <nc>that</nc> is statistically correlated to <nc>the probability</nc> that <nc>a direct citation</nc> exists between <nc>each corresponding pair</nc> of <nc>citationally-related potential input document</nc> and <nc>potential output document</nc> and wherein said <nc>first numerical score</nc> is calculated based at least in <nc>part</nc> on <nc>how many indirect citations</nc> exist between <nc>each</nc> said <nc>pair</nc> of <nc>citationally related documents</nc> and, for <nc>each indirect citation</nc>, <nc>how many citation links</nc> separate <nc>each</nc> said pair of <nc>citationally-related documents</nc>; storing said <nc>first numerical score</nc> for <nc>each</nc> said <nc>pair</nc> of <nc>citationally related documents</nc> on <nc>said computer-readable storage device</nc> in <nc>association</nc> with <nc>said search index</nc>; receiving <nc>a search query</nc> comprising <nc>a third set</nc> of <nc>identification information</nc> identifying <nc>one or more input documents</nc> selected from <nc>said source documents</nc>; using said <nc>third set</nc> of <nc>identification information</nc> to ascertain from <nc>said search index</nc>, via said <nc>one or more computer-processors</nc>, <nc>a fourth set</nc> of <nc>identification information</nc> identifying, for <nc>each</nc> of said <nc>one or more input documents</nc>, <nc>a selected number</nc> of <nc>corresponding output documents</nc> and, for <nc>each pair</nc> of <nc>input document</nc> and <nc>corresponding output document</nc>, said <nc>first numerical score</nc>; calculating, responsive to receiving said <nc>search query</nc>, via said <nc>one or more computer-processors</nc>, <nc>a second numerical score</nc> <nc>that</nc> is statistically correlated to <nc>the probability</nc> that <nc>a direct citation</nc> exists between any of said <nc>one or more input documents</nc> and <nc>each</nc> of said corresponding output documents, and wherein said <nc>second numerical score</nc> is calculated based at least in <nc>part</nc> on said <nc>first numerical score</nc>; generating, via said <nc>one or more computer-processors</nc>, <nc>a search query</nc> result set comprising <nc>identification information</nc> identifying one or more of said <nc>output documents</nc> and wherein said <nc>search query</nc> result <nc>set</nc> is sorted or ranked in <nc>accordance</nc> with <nc>said second numerical score</nc>; and storing said <nc>search query</nc> result set on <nc>said computer-readable storage device</nc>.
| 6
|
6. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 further comprising visually displaying said <nc>search query</nc> result set in <nc>the form</nc> of <nc>an interactive chart</nc>, <nc>graph</nc>, or <nc>map</nc>.
|
9392008
|
14923364
|
1. A system for determining <nc>a merchant breach</nc>, <nc>the system</nc> comprising: <nc>a memory device</nc> configured to store <nc>a set</nc> of <nc>instructions</nc>; and <nc>one or more processors</nc> configured to execute <nc>the set</nc> of <nc>instructions</nc> <nc>that</nc> cause <nc>the one or more processors</nc> to: acquire <nc>card transaction data</nc> from <nc>one or more merchants</nc>, wherein <nc>the acquisition</nc> of <nc>the card transaction data</nc> is performed at <nc>a first predetermined periodic interval</nc>; <nc>store information</nc> relating to <nc>payment cards</nc> associated with <nc>the card transaction data</nc>; for <nc>at least some</nc> of <nc>the payment cards</nc>, automatically determine <nc>at least one value</nc> indicative of <nc>card health</nc> and store <nc>the determined values</nc> for <nc>the payment cards</nc>, wherein <nc>determination</nc> of <nc>the at least one value</nc> indicative of <nc>card health</nc> is performed at <nc>a second predetermined periodic interval</nc>; accumulate, over <nc>a period</nc> of <nc>time</nc>, <nc>card health data</nc> for <nc>the payment cards</nc> based on <nc>the determined at least one value</nc> indicative of <nc>card health</nc>; store <nc>the accumulated card health data</nc>; identify <nc>a potential merchant breach</nc> and <nc>the timing</nc> of <nc>the potential merchant breach</nc> based on <nc>a comparison</nc> between <nc>the accumulated card health data</nc> and <nc>the stored information</nc> relating to <nc>payment cards</nc> associated with <nc>the card transaction data</nc>; and display in <nc>a graph</nc> of <nc>a graphical user interface</nc> of <nc>the system</nc>, <nc>at least a portion</nc> of <nc>the accumulated card health data</nc> and <nc>the card transaction data</nc> indicating <nc>the potential merchant breach</nc>.
| 4
|
4. <nc>The system</nc> of <nc>claim</nc> 1 , wherein <nc>the card health</nc> is re-determined at least once for <nc>each</nc> of <nc>the payment cards</nc>.
|
8595186
|
12135089
|
1. A mobile device, comprising: <nc>at least one processor</nc>; memory storing <nc>instructions</nc>, <nc>the instructions</nc> comprising: <nc>instructions</nc> for <nc>a declaratory markup language renderer</nc> configured to instruct <nc>the at least one processor</nc> to render <nc>a declaratory markup language component</nc> of <nc>a widget application</nc> on <nc>a display</nc> of <nc>the mobile device</nc>; <nc>instructions</nc> for <nc>a compiled programming language execution engine</nc> configured to instruct <nc>the at least one processor</nc> to execute <nc>a compiled programming language component</nc> of <nc>a widget application</nc> installed on <nc>the mobile device</nc>; <nc>instructions</nc> for <nc>a mobile device API</nc>, adapted to be accessible to <nc>a widget application</nc>, and providing <nc>access</nc> to <nc>a device service API</nc> of <nc>the mobile device</nc>; and <nc>instructions</nc> for <nc>a widget manager</nc> configured to instruct <nc>the at least one processor</nc> to: crawl <nc>one or more remote network resources</nc> accessible via <nc>a network</nc> for <nc>widget applications</nc>; to automatically determine one or more of <nc>the widget applications</nc> for <nc>download</nc> based on <nc>a user profile</nc> associated with <nc>the mobile device</nc>; to automatically download <nc>the one or more widget applications</nc>, from <nc>a remote network location</nc> to <nc>the mobile device</nc>, to constitute <nc>a set</nc> of <nc>downloaded widget applications</nc>; and to install <nc>the set</nc> of <nc>downloaded widget applications</nc>, wherein the downloading and installing are based on <nc>the user profile</nc> associated with <nc>the mobile device</nc>, without <nc>user interaction</nc> with <nc>the mobile device</nc>; wherein <nc>the set</nc> of <nc>downloaded widget applications</nc> corresponds to <nc>a first set</nc> of <nc>widget applications</nc>; and <nc>instructions</nc> for <nc>the widget manager</nc> configured to instruct <nc>the at least one processor</nc> to automatically uninstall one or <nc>both</nc> of <nc>the first set</nc> of <nc>widget applications</nc> and a second set of <nc>widget applications</nc> on <nc>the mobile device</nc> based at least in <nc>part</nc> on <nc>user preferences</nc>, without <nc>user interaction</nc> with <nc>the mobile device</nc>.
| 29
|
29. <nc>The mobile device</nc> of <nc>claim</nc> 1 , wherein <nc>the device service API</nc> is <nc>an API</nc> for <nc>a software program</nc> for storing and managing <nc>personal contacts</nc>.
|
7840400
|
11562142
|
1. <nc>A computer-implemented method</nc> for causing <nc>a computer</nc> to understand <nc>a natural language text</nc>, <nc>the method</nc> comprising causing <nc>the computer</nc> to execute <nc>steps</nc> of: receiving <nc>a natural language text</nc>; extracting <nc>at least one parameter value</nc> from <nc>said natural language text</nc> or <nc>a form</nc> thereof; identifying <nc>at least one parameter type</nc> related to <nc>each extracted parameter value</nc>; providing <nc>at least one restatement</nc> of <nc>said natural language text</nc>, <nc>each at least one restatement</nc> having, embedded within, <nc>at least one of said identified parameter types</nc>; extracting <nc>at least one overall category value</nc> from said <nc>at least one restatement</nc> or <nc>a form</nc> thereof; selecting <nc>a subcategory extractor</nc> corresponding to one of said extracted <nc>at least one overall category</nc>, and using said <nc>selected subcategory extractor</nc> to extract <nc>at least one subcategory value</nc>; choosing one of said <nc>at least one extracted subcategory values</nc>; and evaluating said <nc>at least one identified parameter type</nc> in <nc>relation</nc> to said <nc>chosen subcategory value</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: if <nc>at least one parameter value</nc> is extracted <nc>which</nc> corresponds to <nc>a parameter type</nc> of <nc>which</nc> <nc>at least two values</nc> are defined for said <nc>chosen subcategory value</nc>, evaluating <nc>a relationship</nc> between said <nc>at least one extracted parameter value</nc> and said <nc>at least two values</nc> defined for <nc>said subcategory value</nc>.
|
9785638
|
14883444
|
1. A method comprising: storing, by <nc>a construction document storage and search system</nc>, construction project specification documents in <nc>a data storage system</nc>, wherein <nc>the construction project specification documents</nc> are formatted to have <nc>a predefined uniform organizational structure</nc> according to <nc>a standard</nc> for organizing <nc>construction specification documents</nc>; receiving, by <nc>the construction document storage and search system</nc>, <nc>a search query</nc> comprising <nc>a first search criteria</nc> from <nc>a user</nc> electronically via <nc>a graphical user interface</nc>; analyzing, by <nc>the construction document storage and search system</nc>, the construction project specification documents to determine <nc>a first number</nc> of <nc>documents</nc> <nc>that</nc> satisfy <nc>the first search criteria</nc>; responsive to <nc>the search query</nc>, generating, by <nc>the construction document storage and search system</nc>, <nc>a display</nc> reflecting <nc>data</nc> regarding <nc>the number</nc> of <nc>documents</nc> <nc>that</nc> satisfy <nc>the first search criteria</nc>, <nc>the display</nc> includes <nc>a plurality</nc> of <nc>charts</nc> displaying <nc>information</nc> relating to <nc>the first number</nc> of <nc>documents</nc>, wherein <nc>a first chart</nc> of <nc>the plurality</nc> of <nc>charts</nc> includes <nc>information</nc> relating to <nc>a second search criteria</nc> different than <nc>the first search criteria</nc>, and wherein <nc>the information</nc> relating to one of <nc>the first search criteria</nc> or <nc>the second search criteria</nc> of <nc>the first chart</nc> is selectable by <nc>the user</nc> to provide <nc>a modified search criteria</nc> configured to cause <nc>the plurality</nc> of <nc>charts</nc> to change in <nc>response</nc> to <nc>the modified search criteria</nc> based on <nc>the construction project specification documents</nc>.
| 2
|
2. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the search query</nc> further comprises <nc>the second search criteria</nc>, and the method further comprising: analyzing, by <nc>the construction document storage system</nc>, <nc>the construction project specification documents</nc> to determine <nc>a second number</nc> of <nc>documents</nc> <nc>that</nc> satisfy <nc>the second search criteria</nc>, wherein <nc>the plurality</nc> of <nc>charts</nc> includes <nc>a comparison chart</nc> <nc>that</nc> compare <nc>the first number</nc> of <nc>documents</nc> to <nc>the second number</nc> of <nc>documents</nc>.
|
9582492
|
14644291
|
1. <nc>A method</nc> comprising: receiving <nc>a candidate lexical kernel unit</nc> comprising <nc>a word token sequence</nc> <nc>that</nc> includes <nc>two or more words</nc>; retrieving <nc>domain terms</nc> <nc>that</nc> contain <nc>the two or more words</nc> from <nc>a terminology resource file</nc> of <nc>domain terms</nc> associated with <nc>a domain</nc>; analyzing <nc>the candidate lexical kernel unit</nc> and <nc>the retrieved domain terms</nc> to determine whether <nc>the candidate lexical kernel unit</nc> satisfies <nc>specified criteria</nc> for <nc>use</nc> as <nc>a building block</nc> by <nc>a natural-language processing</nc> <nc>(NLP) tool</nc> for building <nc>larger lexical units</nc> in <nc>the domain</nc>, <nc>each</nc> of <nc>the larger lexical units</nc> including <nc>a greater number</nc> of <nc>words</nc> than <nc>the candidate lexical kernel unit</nc>; identifying <nc>the candidate lexical kernel unit</nc> as <nc>a lexical kernel unit</nc> based on determining that <nc>the candidate lexical kernel unit</nc> satisfies <nc>the specified criteria</nc>; and outputting <nc>the lexical kernel unit</nc> to <nc>a domain-specific lexical kernel unit file</nc> for <nc>input</nc> to <nc>the NLP tool</nc> for <nc>use</nc> as <nc>a lexical resource</nc> in parsing <nc>natural language text</nc> in <nc>the domain</nc>, <nc>the parsing</nc> including identifying <nc>domain-specific terms</nc> in <nc>the natural language text</nc> in <nc>the domain</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein at least two of <nc>the two or more words</nc> included in <nc>the candidate lexical kernel unit</nc> are not contiguous in least one of <nc>the retrieved domain terms</nc>.
|
9900297
|
14059268
|
1. A system comprising: <nc>one or more processors</nc> configurable to cause: displaying <nc>content items</nc> in <nc>a web browser</nc>, <nc>each content item</nc> having <nc>content</nc> displayed according to <nc>one or more formatting tags</nc>; identifying a first one of <nc>the displayed content items</nc> having <nc>first content</nc> displayed according to <nc>at least a first formatting tag</nc>, <nc>the first content item</nc> having <nc>dimensions</nc> comprising <nc>a height</nc> and <nc>a width</nc>; identifying <nc>access credentials</nc> for accessing <nc>one or more content</nc> posting <nc>services</nc>, <nc>the access credentials</nc> comprising <nc>a user name</nc> and <nc>a first network address</nc>; accessing <nc>the one or more content</nc> posting <nc>services</nc> using <nc>the access credentials</nc>; processing <nc>publication information</nc> associated with <nc>the one or more content posting services</nc>, <nc>the publication information</nc> comprising <nc>an identifier</nc> and <nc>a second network address</nc>; determining that <nc>the second network address</nc> of <nc>the publication information</nc> matches <nc>the first network address</nc> of <nc>the access credentials</nc>; selecting, responsive to determining that <nc>the second network address</nc> matches <nc>the first network address</nc>, <nc>the identifier</nc> of <nc>the publication information</nc>; generating <nc>a post</nc> including <nc>the first content item</nc> according to: <nc>the selected identifier</nc>, and <nc>the dimensions</nc>; displaying <nc>the post</nc> in <nc>a preview interface</nc>, <nc>the preview interface</nc> enabling: <nc>user entry</nc>, <nc>user modification</nc> and <nc>user deletion</nc> of <nc>content</nc> in <nc>the post</nc> including <nc>the first content item</nc>, and <nc>user selection</nc> of <nc>each</nc> of <nc>the one or more content posting services</nc>; and providing <nc>the post</nc> for posting on <nc>the one or more content</nc> posting <nc>services</nc>.
| 3
|
3. <nc>The system</nc> of <nc>claim</nc> 1 , wherein <nc>the post</nc> comprises a Hypertext Markup Language (<nc>HTML</nc>) description.
|
8566790
|
13093992
|
1. <nc>A system</nc> comprising: <nc>a processor</nc>; <nc>a storage medium</nc>; <nc>a data representation language schema</nc> stored on <nc>the storage medium</nc>; and <nc>a script</nc> for manipulating <nc>a data representation language document</nc> in <nc>accordance</nc> with <nc>a scripting language</nc> stored on <nc>the storage medium</nc>; and <nc>an interpreter</nc> to evaluate <nc>the script</nc>, wherein <nc>the processor</nc> is configured to import <nc>the data representation language schema</nc> using <nc>an import function</nc> of <nc>the script</nc>; wherein <nc>the processor</nc> is further configured to create <nc>one or more new types</nc> within <nc>the scripting language</nc> for manipulating <nc>data representation language values</nc> conforming to <nc>one or more type definitions</nc> within <nc>the data representation language schema</nc>; and wherein <nc>the processor</nc> is further configured to interpret <nc>the script</nc> using <nc>the one or more new types</nc>.
| 5
|
5. <nc>The system</nc> of <nc>claim</nc> 1 , wherein <nc>the script</nc> includes <nc>a validate method</nc> to check if <nc>a data representation language schema data constraint</nc> is violated.
|
8245128
|
10197738
|
1. <nc>A method</nc> of operating <nc>an application</nc> on <nc>a client mobile computing device</nc>, comprising: receiving on <nc>the client mobile computing device</nc>: <nc>a database snapshot</nc> <nc>which</nc> describes <nc>a data</nc> set in <nc>a remote database</nc>; and <nc>a presentation format</nc> for <nc>a first application page</nc> <nc>that</nc> describes how <nc>the content</nc> of <nc>the database snapshot</nc> is presented; receiving at <nc>a client agent</nc> residing on <nc>the client mobile computing device</nc> <nc>a request</nc> from <nc>a client browser</nc> for <nc>the first application page</nc>; using <nc>a push listener</nc> of <nc>the client agent</nc> to receive <nc>a message</nc> that <nc>the server</nc> pushes to <nc>the client mobile computing device</nc>, wherein <nc>the pushed message</nc> includes <nc>a SQL statement</nc> <nc>which</nc> is executed by <nc>the client mobile computing device</nc> to update <nc>the database snapshot</nc>; producing <nc>the first application page</nc> at <nc>the mobile computing device</nc> offline by applying <nc>a script engine</nc> of <nc>the client agent</nc> to <nc>the presentation format</nc>; and presenting <nc>the produced application page</nc> at <nc>the mobile computing device</nc> in <nc>response</nc> to <nc>the request</nc>.
| 11
|
11. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: receiving from <nc>a request</nc> for <nc>a second page</nc> of <nc>the application</nc>; determining that <nc>a presentation format</nc> of <nc>the second page</nc> is not available on <nc>the mobile computing device</nc>; and queuing <nc>the request</nc> for <nc>the second page</nc> for <nc>transmission</nc> to <nc>a remote computer system</nc>.
|
5469354
|
07843162
|
1. <nc>A document data processing method</nc> for retrieving <nc>a document</nc> containing <nc>at least a search term</nc> designated by <nc>an operator</nc> from <nc>a document database</nc> registering therein <nc>document information</nc> in <nc>terms</nc> of <nc>character code data</nc> while referring to <nc>textual content</nc> of <nc>said document</nc>, comprising <nc>steps</nc> of: upon <nc>registration</nc> of <nc>text documents</nc> in said document database, creating <nc>condensed texts</nc> by decomposing <nc>each</nc> of <nc>textual character strings</nc> of <nc>the documents</nc> to be registered into <nc>fragmental character strings</nc> on <nc>the basis</nc> of at least one of <nc>character species</nc> including <nc>katakana character</nc>, <nc>hiragana character</nc>, <nc>kanji character</nc>, <nc>alphabetic character</nc>, <nc>numeric character</nc>, and <nc>symbol character</nc> and checking <nc>mutual inclusion relations</nc> possibly existing among said <nc>fragmental character strings</nc> resulting from <nc>said decomposition</nc>, to thereby create <nc>the condensed texts</nc> <nc>each</nc> constituted by <nc>a set</nc> of <nc>the fragmental character strings</nc> in <nc>which</nc> <nc>any character string</nc> found to be included by <nc>other character string</nc> is eliminated; creating <nc>a component character table</nc> in <nc>which</nc> <nc>characters</nc> occurring in <nc>each</nc> of <nc>said condensed texts</nc> are registered without <nc>duplication</nc>; and registering in said <nc>document database</nc> said <nc>condensed texts</nc> together with <nc>said component character table</nc> in <nc>addition</nc> to <nc>the texts</nc> of <nc>the document</nc> to be registered; and upon <nc>retrieval</nc> of <nc>the document</nc> containing <nc>the designated search term</nc>, executing first <nc>a component character table search</nc> for thereby extracting <nc>those documents</nc> <nc>which</nc> contain <nc>all species</nc> of <nc>characters</nc> constituting <nc>the search term</nc> designated by <nc>the operator</nc> by consulting said <nc>component character table</nc>; executing subsequently <nc>a condensed text search</nc> by consulting <nc>the condensed texts</nc> of <nc>the documents</nc> extracted through said <nc>component character table search</nc> for extracting <nc>only the documents</nc> corresponding to <nc>the condensed texts</nc> <nc>which</nc> contain <nc>the fragmental character strings</nc> constituting <nc>the search term</nc> designated by <nc>the operator</nc> to thereby select <nc>the documents</nc> containing <nc>the designated search term</nc>; and executing finally <nc>a text body search</nc> for extracting <nc>a document</nc> <nc>which</nc> satisfies <nc>query condition</nc> imposed on <nc>the search term</nc> by consulting <nc>the texts</nc> of <nc>the documents</nc> extracted through said <nc>component character table search</nc> and said <nc>condensed text search</nc>.
| 2
|
2. <nc>A document data processing method</nc> for <nc>document retrieval</nc> according to <nc>claim</nc> 1, wherein said <nc>component character table</nc> registers without <nc>duplication</nc> <nc>all the characters</nc> as used on <nc>a document basis</nc>.
|
9691068
|
13327140
|
1. <nc>A method</nc> under <nc>control</nc> of <nc>one or more computing systems</nc> configured with <nc>specific executable instructions</nc>, <nc>the method</nc> comprising: obtaining <nc>information</nc> about <nc>a work</nc> from <nc>one or more network resources</nc>; deriving <nc>metadata</nc> of <nc>the work</nc> from <nc>the information</nc> obtained about <nc>the work</nc>; determining, based at least in <nc>part</nc> on <nc>copyright laws</nc> of <nc>a country</nc>, that <nc>a subset</nc> of <nc>the metadata</nc> has <nc>a minimum amount</nc> of <nc>information</nc> to make <nc>a determination</nc> related to <nc>the work</nc> being in <nc>the public domain</nc> for <nc>the country</nc>; generating, based at least partly on <nc>the subset</nc> of <nc>the metadata</nc>, <nc>a numerical confidence level</nc> indicative of <nc>the work</nc> being in <nc>the public domain</nc> for <nc>the country</nc> or <nc>the work</nc> being protected by <nc>copyright</nc> in <nc>the country</nc>; comparing <nc>the numerical confidence level</nc> to <nc>a predetermined threshold value</nc> for <nc>the country</nc>; determining that <nc>the numerical confidence level</nc> is above <nc>the predetermined threshold value</nc>; determining, based at least partly on <nc>the numerical confidence level</nc> being above <nc>the predetermined threshold value</nc>, <nc>a recommendation</nc> to make <nc>the work</nc> available as <nc>a public domain document</nc> in <nc>the country</nc>; and generating <nc>a user interface</nc> <nc>that</nc> includes <nc>the numerical confidence level</nc> and <nc>the recommendation</nc>.
| 3
|
3. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the user interface</nc> includes <nc>a selectable option</nc> to perform <nc>an action</nc> corresponding to <nc>the recommendation</nc>.
|
6044347
|
08906416
|
1. <nc>A method</nc> for <nc>use</nc> in <nc>a processing system</nc> for managing <nc>a dialogue</nc> between <nc>the system</nc> and <nc>a user</nc>, <nc>the method</nc> comprising <nc>the steps</nc> of: receiving <nc>an input signal</nc> representing one of <nc>speech</nc> and <nc>text</nc> associated with <nc>at least a portion</nc> of <nc>the dialogue</nc>; processing <nc>a set</nc> of <nc>frames</nc> characterizing <nc>a subject</nc> of <nc>the dialogue</nc>, associated with <nc>the input signal</nc>, each frame including <nc>one or more properties</nc> <nc>that</nc> describe <nc>a corresponding object</nc> <nc>which</nc> may be referenced during <nc>the dialogue</nc>, wherein <nc>a weight</nc> is assigned to <nc>each</nc> of <nc>at least a subset</nc> of <nc>the properties</nc> represented by <nc>the set</nc> of <nc>frames</nc>, such that <nc>the assigned weights</nc> indicate <nc>the relative importance</nc> of <nc>the corresponding properties</nc>; utilizing <nc>the weights</nc> to determine <nc>which</nc> of <nc>a plurality</nc> of <nc>possible responses</nc> <nc>the system</nc> should generate for <nc>a given user input</nc> during <nc>the dialogue</nc>; and generating <nc>an output signal</nc> for <nc>sensory perception</nc> by <nc>the user</nc> representing at least one of <nc>the plurality</nc> of <nc>possible responses</nc>.
| 9
|
9. <nc>The method</nc> of <nc>claim</nc> 1 further including <nc>the steps</nc> of identifying a particular one of <nc>a plurality</nc> of <nc>dialogue motivators</nc> associated with <nc>the given user input</nc>, and utilizing <nc>the identified dialogue motivator</nc> in <nc>conjunction</nc> with <nc>the weights</nc> to determine <nc>which</nc> of <nc>a plurality</nc> of <nc>possible responses</nc> <nc>the system</nc> should generate for <nc>a given user input</nc> during <nc>the dialogue</nc>.
|
8078629
|
12578339
|
1. <nc>A computer program product</nc> stored on <nc>one or more non-transitory computer readable storage media</nc> and comprising <nc>instructions</nc> <nc>that</nc>, when executed, cause <nc>an apparatus</nc> to: determine, for <nc>a document</nc> <nc>that</nc> contains <nc>a first phrase</nc>, <nc>an expected number</nc> of <nc>related phrases</nc> <nc>that</nc> are related to <nc>the first phrase</nc> and are expected to be present in <nc>the document</nc>; determine for <nc>the document</nc>, and for <nc>the first phrase</nc> in <nc>the document</nc>, <nc>an actual number</nc> of <nc>related phrases</nc> present in <nc>the document</nc>; and identify <nc>the document</nc> as <nc>a spam document</nc> by comparing <nc>the actual number</nc> of <nc>related phrases</nc> present in <nc>the document</nc> with <nc>the expected number</nc> of <nc>related phrases</nc>, wherein determining <nc>the expected number</nc> of <nc>related phrases</nc> includes: traversing <nc>an index</nc> of <nc>a plurality</nc> of <nc>documents</nc>; for <nc>each</nc> of <nc>the indexed documents</nc>, determining <nc>a set</nc> of <nc>phrases</nc> in <nc>the document</nc>, and for <nc>each phrase</nc> in <nc>the set</nc>, determining <nc>a number</nc> of <nc>related phrases</nc> also in <nc>the document</nc>; and determining <nc>the expected number</nc> of <nc>related phrases</nc> based on <nc>the determined number</nc> of <nc>related phrases</nc> across <nc>the traversed documents</nc>.
| 4
|
4. <nc>The computer program product</nc> of <nc>claim</nc> 1 , wherein <nc>the determination</nc> of <nc>the number</nc> of <nc>the related phrases</nc> expected to be present in <nc>the document</nc> is based on <nc>a statistical analysis</nc> of <nc>a plurality</nc> of <nc>documents</nc> <nc>that</nc> include <nc>the first phrase</nc> and <nc>related phrases</nc>, and wherein identifying <nc>the document</nc> as <nc>a spam document</nc>, further comprises: determining <nc>a standard deviation</nc> of <nc>the expected number</nc> of <nc>related phrases</nc>; and responsive to <nc>the actual number</nc> of <nc>related phrases</nc> present in <nc>the document</nc> exceeding <nc>the expected number</nc> of <nc>related phrases</nc> by <nc>at least a multiple</nc> of <nc>a standard deviation</nc> of <nc>the expected number</nc> of <nc>related phrases</nc>, identifying <nc>the document</nc> as <nc>a spam document</nc>.
|
9740692
|
14533530
|
1. <nc>A method</nc> for creating <nc>a flexible structure description</nc>, <nc>the method</nc> comprising: receiving <nc>an image</nc> of <nc>a document</nc> of <nc>a particular document type</nc> <nc>that</nc> contains <nc>a table</nc>; receiving <nc>an entry</nc> describing <nc>an item</nc> in <nc>the table</nc>; searching for <nc>title elements</nc> based upon <nc>the entry</nc>; detecting <nc>data fields</nc> and <nc>anchor elements</nc> for <nc>the entry</nc>; generating, using <nc>a processor</nc>, <nc>a flexible structure description</nc> for <nc>the particular document type</nc> <nc>that</nc> includes <nc>a set</nc> of <nc>search elements</nc> for <nc>each data field</nc> in <nc>the image</nc> of <nc>the document</nc> and <nc>the title elements</nc>; matching <nc>the flexible structure description</nc> against <nc>the image</nc>; and extracting <nc>data</nc> from <nc>the image</nc> based upon <nc>the matching</nc> of <nc>the flexible structure description</nc> against <nc>the image</nc>.
| 2
|
2. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising adjusting, using <nc>the processor</nc>, <nc>the flexible structure description</nc> based on <nc>user corrections</nc> of <nc>detected data fields</nc>, <nc>title elements</nc>, and/or <nc>anchor elements</nc>.
|
10134050
|
14699072
|
1. <nc>A method</nc> for facilitating <nc>use</nc> of <nc>mobile devices</nc> to respond to <nc>questions</nc> submitted to <nc>a question</nc> and <nc>answer customer support system</nc>, to improve <nc>a rate</nc> of <nc>response</nc> to <nc>the questions</nc> by <nc>customer support personnel</nc>, <nc>the method</nc> comprising: <nc>training</nc>, using <nc>a computing system</nc>, <nc>predictive models</nc> using <nc>historical question</nc> and answer <nc>data</nc>, <nc>the training</nc> resulting in <nc>at least one predictive model</nc> configured to estimate, based on <nc>a received question</nc>, <nc>an expected answer length</nc>, determine whether <nc>an answer</nc> to <nc>the question</nc> is likely to include <nc>a web link</nc>, and determine whether <nc>a question answerer</nc> is more likely than not to have to perform <nc>research</nc> in <nc>order</nc> to answer <nc>the question</nc>; receiving, with <nc>a computing system</nc> having <nc>a processor</nc> and <nc>a memory</nc>, <nc>a first question</nc> from <nc>a user</nc> having <nc>a first type</nc>; determining that <nc>the received question</nc> is <nc>a first type</nc> of <nc>question</nc> and forming and sending <nc>a response</nc> to <nc>the user</nc> with <nc>recommendations</nc> for <nc>the user</nc> to reform <nc>the question</nc> into <nc>a different type</nc>; receiving <nc>a reformed first question</nc> from <nc>the user</nc>, <nc>the reformed question</nc> being <nc>the first question</nc> transformed into <nc>a second type</nc> of <nc>question</nc>; analyzing, using <nc>the one or more predictive models</nc>, the reformed first question with <nc>a question</nc> and <nc>answer customer support system</nc> of <nc>the computing system</nc> by determining <nc>one or more attributes</nc> of <nc>the reformed first question</nc> and determining that <nc>the reformed first question</nc> is <nc>a mobile device answerable question</nc>, because one or more of <nc>the determined one or more attributes</nc> of <nc>the reformed first question</nc> satisfy <nc>one or more mobile device question criteria</nc>; prioritizing, using <nc>the computing system</nc>, the answering of <nc>the reformed first question</nc> over <nc>the answering</nc> of <nc>questions</nc> <nc>that</nc> are not <nc>mobile device</nc> <nc>answerable questions</nc>; configuring <nc>multiple user interface elements</nc>, at least partially based on <nc>the one or more attributes</nc> of <nc>the reformed first question</nc> by at least prepopulating <nc>the user interface elements</nc> to include <nc>at least one proposed answer</nc> to <nc>the reformed first question</nc>; and providing <nc>the user interface elements</nc> with <nc>the question and answer customer support system</nc> for <nc>display</nc> in <nc>a user interface</nc> on <nc>the mobile device</nc> of <nc>a first question answerer</nc>.
| 7
|
7. <nc>The method</nc> of <nc>claim</nc> 1 , wherein configuring <nc>the multiple user interface elements</nc> includes <nc>pre-populating response buttons</nc> within <nc>the user interface</nc> with <nc>potential answers</nc> to <nc>the question</nc> to enable <nc>the one or more support personnel</nc> to generate <nc>the response</nc> by selecting one or more of <nc>the response buttons</nc>.
|
7912711
|
11903550
|
1. <nc>A data processing device</nc> for generating, from <nc>a preset code</nc>, <nc>filter data</nc> to be afforded to <nc>a speech synthesis filter</nc> adapted for synthesizing <nc>the speech</nc> based on <nc>linear prediction coefficients</nc> and <nc>a preset input signal</nc>, comprising: <nc>code decoding means</nc> for decoding said code produced by encoding <nc>original filter data</nc>, to <nc>output</nc> decoded <nc>filter data</nc>; <nc>acquisition</nc> means for acquiring <nc>preset tap coefficients</nc> as found by carrying out <nc>learning</nc>, wherein said <nc>tap coefficients</nc> are used to predict <nc>the original filter data</nc> from said decoded <nc>filter data</nc>; and <nc>prediction</nc> means for carrying out <nc>preset predictive calculations</nc>, using said <nc>tap coefficients</nc> and <nc>the decoded filter data</nc>, to find <nc>prediction values</nc> of <nc>said filter data</nc>, to send <nc>the so found prediction values</nc> to said <nc>speech synthesis filter</nc> for <nc>use</nc> as <nc>linear prediction coefficients</nc> in said <nc>speech syntheses filter</nc>.
| 10
|
10. <nc>The data processing device</nc> according <nc>to claim</nc> 1 further comprising: said <nc>speech synthesis filter</nc>.
|
9043899
|
14084270
|
1. <nc>A method</nc> of providing <nc>controlled, electronic access</nc> to <nc>variable domain data</nc> stored in <nc>a data processing system</nc>, <nc>the method</nc> comprising: performing by <nc>the data processing system</nc> programmed with <nc>code</nc> stored in <nc>a memory</nc> and executing by <nc>a processor</nc> of <nc>the data processor system</nc> to configure <nc>the data processing system</nc> into <nc>a machine</nc>: receiving <nc>information</nc> from <nc>a computer system</nc> of <nc>a principal</nc> <nc>that</nc> includes <nc>information</nc> identifying <nc>the principal</nc>; storing at least <nc>the information</nc> identifying <nc>the principal</nc>; accessing <nc>a data security model</nc> and <nc>a variable domain data model</nc> from <nc>the memory</nc>; performing <nc>one or more logical relationship operations</nc> on <nc>a data security model</nc> and <nc>a variable domain data model</nc> using <nc>security attributes</nc> of <nc>the data security model</nc> to determine <nc>a level</nc> of <nc>access</nc> to <nc>resource data</nc> in <nc>the variable domain data model</nc> to be granted to <nc>the principal</nc>, wherein <nc>the data security model</nc> and <nc>the variable domain model</nc> share <nc>a common logical relationship data structure</nc>, and <nc>the logical data relationship structure</nc> includes <nc>logical relationship expressions</nc> <nc>that</nc> define <nc>data</nc> within <nc>the data security model</nc> and <nc>the variable domain model</nc>; and granting <nc>the principal access</nc> via <nc>the computer system</nc> of <nc>the principal</nc> to <nc>the resource data</nc> in <nc>accordance</nc> with <nc>the determined level</nc> of <nc>resource data access</nc> to be granted to <nc>the principal</nc>, wherein <nc>the principal</nc> comprises <nc>an entity</nc> <nc>that</nc> has controlled <nc>access</nc> to <nc>the resource data</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 wherein <nc>the level</nc> of <nc>resource data access</nc> granted to <nc>the principal</nc> comprises <nc>rights</nc> to view <nc>the resource data</nc>.
|
9417709
|
14857146
|
1. <nc>A system</nc> for implementing <nc>a sliding input</nc> of <nc>a text</nc> based upon an on-<nc>screen</nc> soft keyboard on <nc>an electronic equipment</nc>, characterized in <nc>that</nc>, said <nc>system</nc> comprises: <nc>a memory device</nc> configured to store <nc>ideal sliding trajectory features</nc> for <nc>words</nc>; and <nc>a processor</nc> coupled to <nc>the memory device</nc>, <nc>the processor</nc> being configured to: <nc>record user-sliding trajectories</nc> and convert <nc>the recorded user-sliding trajectories</nc> into <nc>a user-sliding trajectory feature</nc> set to be processed; filter in <nc>the memory device</nc> and originally choose <nc>the words</nc>, wherein <nc>each</nc> of <nc>the originally chosen words</nc> has similar ideal sliding trajectory features with the user-sliding trajectory feature set; calculate <nc>a similarity</nc> between <nc>the ideal sliding trajectory features</nc> of <nc>each originally chosen word</nc> and said <nc>user-sliding trajectory features</nc> set according to <nc>key points</nc> on <nc>said trajectory</nc>, characterized in <nc>that</nc>, calculating said <nc>similarity</nc> including <nc>the following steps</nc>: calculating <nc>a rough similarity</nc> between <nc>the ideal sliding trajectory features</nc> of <nc>each originally chosen word</nc> and said <nc>user-sliding trajectory features</nc> set; calculating <nc>an accurate similarity</nc> between <nc>the ideal sliding trajectory features</nc> of <nc>each word</nc> obtained from <nc>the rough similarity calculation result</nc> and said <nc>user-sliding trajectory features</nc> set; wherein calculating said <nc>accurate similarity</nc> including calculating <nc>an accurate linear matching distance</nc> between <nc>the ideal sliding trajectory feature</nc> of <nc>each word</nc> obtained from <nc>the rough similarity calculation result</nc> and said user-sliding trajectory feature set; obtain <nc>candidate words</nc> according to <nc>the similarity</nc>, wherein <nc>the ideal sliding trajectory</nc> of <nc>each candidate word</nc> contains at least one of <nc>the key points</nc> or at least one of <nc>the surrounding points</nc> of at least one of <nc>the key points</nc> on <nc>said user-sliding trajectory</nc>; <nc>display</nc> said <nc>candidate words</nc>.
| 8
|
8. <nc>The system</nc> according to <nc>claim</nc> 1 , characterized in <nc>that</nc>, said <nc>processor</nc> is further configured to calculate <nc>a rough similarity</nc> between <nc>the ideal sliding trajectory features</nc> of <nc>each originally chosen word</nc> and said <nc>user-sliding trajectory features</nc> set by <nc>the following steps</nc>: calculating <nc>a linear matching distance</nc> between <nc>the ideal trajectory</nc> of <nc>each originally chosen word</nc> and said user-sliding trajectory feature set.
|
8633932
|
12504263
|
1. <nc>A computer-implemented method</nc> comprising: quantifying <nc>similarities</nc> between <nc>a first multi-frame animation</nc> and <nc>a second multi-frame animation</nc> by determining <nc>a difference</nc> between <nc>content</nc> of <nc>corresponding frame pairs</nc> from <nc>the first multi-frame animation</nc> and <nc>the second multi-frame animation</nc>, and by summing <nc>the frame</nc> <nc>pair differences</nc>; combining <nc>the first and second multi-frame animations</nc> based upon <nc>the quantified similarities</nc> to produce <nc>a combined animation</nc>; compressing <nc>the combined animation</nc> to produce <nc>a first model</nc> <nc>that</nc> represents <nc>the content</nc> of <nc>the first animation</nc> and <nc>the second animation</nc>, wherein <nc>the first model</nc> includes <nc>a finite number</nc> of <nc>shapes</nc>; grouping <nc>the first model</nc> with <nc>a plurality</nc> of <nc>other models</nc> <nc>that</nc> share <nc>a common subset</nc> of <nc>shapes</nc>, wherein <nc>the common subset</nc> of <nc>shapes</nc> provides <nc>approximate renderings</nc> and wherein <nc>a set</nc> of <nc>non-common shapes</nc> provides <nc>detailed renderings</nc>; and initially transmitting <nc>the common set</nc> of <nc>shapes</nc> when <nc>only approximate renderings</nc> of <nc>the first and second multi-frame animations</nc> are needed.
| 2
|
2. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , wherein <nc>the first model</nc> includes <nc>sixteen shapes</nc>.
|
8015129
|
12102453
|
1. <nc>A method</nc> for generating <nc>a parsimonious multi-resolution representation</nc> of <nc>value-item lists</nc>, <nc>the method</nc> comprising: receiving <nc>a set</nc> of <nc>values</nc> for <nc>a set</nc> of <nc>measure attributes</nc> and <nc>a set</nc> of <nc>values</nc> for <nc>a set</nc> of <nc>dimension attributes</nc>; inferring <nc>an initial hierarchic data structure</nc> based at least in <nc>part</nc> on <nc>count data</nc> associated with <nc>the received set</nc> of <nc>values</nc> for <nc>dimension attributes</nc>; distributing <nc>the received set</nc> of <nc>values</nc> for <nc>the set</nc> of <nc>measure</nc> attributes into <nc>a set</nc> of <nc>value-item lists</nc> associated with <nc>a portion</nc> of <nc>leaf nodes</nc> in <nc>the hierarchic data structure</nc>; in <nc>the hierarchic data structure</nc> recursively rearranging from <nc>bottom</nc> to top <nc>the set</nc> of <nc>value-item lists</nc> associated with <nc>the portion</nc> of <nc>leaf nodes</nc> based at least in <nc>part</nc> on <nc>compression performance</nc> stemming from <nc>the rearrangement</nc> of <nc>the set</nc> of <nc>value-item lists</nc>; reducing <nc>a number</nc> of <nc>features</nc> in <nc>the set</nc> of <nc>measure attributes</nc>, prior to <nc>the recursive rearrangement</nc> of <nc>the set</nc> of <nc>value-item lists</nc>; and promoting <nc>a plurality</nc> of <nc>value-item elements</nc> from <nc>the rearranged lists</nc> into <nc>a tree structure</nc> to generate <nc>a parsimonious representation</nc> of <nc>the inferred hierarchic data structure</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein recursively rearranging <nc>the set</nc> of <nc>value-item lists</nc> includes <nc>pairing lists</nc> based at least in <nc>part</nc> on <nc>a stochastic greedy matching</nc>.
|
8073331
|
11952001
|
1. <nc>An optical system</nc> for communicating <nc>optical signals</nc> to and from <nc>a subscriber unit</nc>, comprises: <nc>a first set</nc> of <nc>selected wavelengths</nc>; <nc>a second set</nc> of <nc>selected wavelengths</nc> offset from <nc>the first set</nc> of <nc>selected wavelengths</nc>; <nc>a first subsystem</nc> configured to transmit to <nc>the subscriber unit</nc>, <nc>the optical signals</nc> of <nc>one or more wavelengths</nc> from <nc>the first set</nc> of <nc>selected wavelengths</nc> over <nc>an optical fiber</nc>; <nc>a second subsystem</nc> configured to transmit to <nc>the subscriber unit</nc> <nc>the optical signals</nc> of <nc>one wavelength</nc> selected from <nc>the second set</nc> of <nc>selected wavelengths</nc> in <nc>statistical time division multiplexing</nc> over <nc>the optical fiber</nc>; <nc>a third subsystem</nc> at <nc>the subscriber unit</nc> configured to receive <nc>the optical signals</nc> of <nc>the one or more wavelengths</nc> from <nc>the first set</nc> of <nc>selected wavelengths</nc> and <nc>the optical signals</nc> of <nc>one wavelength</nc> from <nc>the second set</nc> of <nc>selected wavelengths</nc> over <nc>the optical fiber</nc>; <nc>a fourth subsystem</nc> at <nc>a head-end</nc> configured to receive <nc>looped back optical signals</nc> of <nc>one wavelength</nc> from <nc>the second set</nc> of <nc>selected wavelengths</nc> from <nc>the subscriber unit</nc> to <nc>the head-end</nc> over <nc>the optical fiber</nc>; a cyclic arrayed waveguide grating router capable of routing <nc>more than one wavelength</nc> to <nc>the subscriber unit</nc>; and a phase modulator, <nc>an intensity modulator</nc>, <nc>an amplifier</nc> and <nc>a looped back configuration</nc> within <nc>the subscriber unit</nc>.
| 18
|
18. <nc>An optical system</nc> as in <nc>claim</nc> 1 , wherein <nc>the subscriber unit</nc> further comprises <nc>a quantum dot-enabled semiconductor optical amplifier</nc> for amplifying <nc>intensity</nc> <nc>modulated optical signals</nc> selected from <nc>one wavelength</nc> from <nc>the second set</nc> of <nc>wavelengths</nc>.
|
8332359
|
12181234
|
1. <nc>A computer-implemented method</nc> of generating <nc>a document revision history</nc> for <nc>versions</nc> of <nc>a document</nc> managed by <nc>a first electronic document management system</nc> (<nc>EDMS</nc>), <nc>the method</nc> comprising, by <nc>a second EDMS</nc>: importing, from <nc>the first EDMS</nc> into <nc>a first location</nc> of <nc>the second EDMS</nc>, metadata describing <nc>a first document revision history</nc> for <nc>the versions</nc> of <nc>the document</nc> managed by <nc>the first EDMS</nc>, wherein <nc>the first document revision history</nc> comprises <nc>metadata attributes</nc> in <nc>a first format</nc> associated with <nc>the first EDMS</nc> and describes <nc>an actual revision history</nc> of <nc>the document</nc>; generating, from <nc>the metadata</nc> describing <nc>the first document revision history</nc>, <nc>a second document revision history</nc> comprising <nc>metadata attributes</nc> in <nc>a second format</nc> associated with <nc>the second EDMS</nc>, wherein <nc>the second document revision history</nc> provides <nc>a mirrored revision history</nc> mirroring <nc>the actual revision history</nc>, and wherein <nc>one or more metadata attributes</nc> in <nc>the second format</nc> are distinct from <nc>one or more corresponding metadata attributes</nc> in <nc>the first format</nc>; importing <nc>the second document revision history</nc> from <nc>the first location</nc> into <nc>a repository</nc> managed by <nc>the second EDMS</nc>; and importing <nc>the versions</nc> of <nc>the document</nc> from <nc>the first location</nc> into <nc>the repository</nc> managed by <nc>the second EDMS</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: prior to generating <nc>the second document revision history</nc>, determining that one or more of <nc>the versions</nc> of <nc>the document</nc> are stored in <nc>the second EDMS</nc> and determining that <nc>the document</nc> has been modified by <nc>users</nc> of <nc>the first EDMS</nc> and <nc>second EDMS</nc> after previously being imported; and when importing <nc>the second document revision history</nc> from <nc>the first location</nc> into <nc>the repository</nc> managed by <nc>the second EDMS</nc>, overwriting <nc>at least a portion</nc> of <nc>an existing revision history</nc> associated with <nc>the one or more versions</nc> of <nc>the document</nc> stored in <nc>the second EDMS</nc>.
|
8280827
|
12064726
|
1. <nc>A multilevel semiotic and fuzzy logic user and metadata interface apparatus</nc> for <nc>varying content manipulation</nc> of <nc>an interactive multimedia representation</nc> in <nc>a multimedia system</nc>, wherein <nc>the multilevel semiotic and fuzzy logic user and metadata interface apparatus</nc> comprises: <nc>(a) an interactive user interface</nc> comprising <nc>i.</nc> <nc>one or more multilevel semiotic means</nc> addressable by <nc>a user</nc>, wherein <nc>the one or more multilevel semiotic means</nc> comprise <nc>an interactive content display</nc>; and <nc>ii</nc>. <nc>a fuzzy logic descriptor</nc> set <nc>module</nc> that <nc>stores multiple fuzzy logic descriptor sets</nc>, wherein each fuzzy logic descriptor set is related to <nc>a respective level</nc> of <nc>each</nc> of <nc>the multilevel semiotic means</nc> and describes <nc>at least one level</nc> of <nc>significance</nc> of <nc>interactivity</nc> corresponding to <nc>the respective level</nc> of <nc>each</nc> of <nc>the multilevel semiotic means</nc>, wherein <nc>the level</nc> of <nc>significance</nc> is directly related to <nc>the importance</nc> of <nc>a level</nc> addressed and thus selected by <nc>the user</nc> of <nc>the multilevel semiotic means</nc>, wherein <nc>each fuzzy logic descriptor set</nc> defines <nc>at least one fuzzy semiotic membership function</nc> definable by and based on addressing of <nc>a level</nc> of <nc>each</nc> of <nc>the multilevel semiotic means</nc> and <nc>at least one significance</nc> attributed to <nc>the level</nc>; (b) a metadata layer <nc>that</nc> links <nc>the one or more multilevel semiotic means</nc> to <nc>interactivity points</nc> present in <nc>the content</nc> of <nc>the interactive multimedia representation</nc>, wherein <nc>the interactivity points</nc> are defined by <nc>the metadata layer</nc> so that <nc>each level</nc> of <nc>each</nc> of <nc>the one or mores multilevel semiotic means</nc> identifies <nc>a first interactivity point</nc> in <nc>the content</nc> and allows, by <nc>selection</nc> of <nc>the first interactivity point</nc>, selectively varying of <nc>content manipulation</nc> and <nc>receipt</nc> of <nc>a resulting interactive presentation</nc> of <nc>content</nc> according to <nc>the selective varying</nc> of <nc>content manipulation</nc>; and <nc>(c) a user control device</nc> operably connected to address <nc>the one or more multilevel semiotic means</nc>.
| 24
|
24. <nc>A multilevel semiotic and fuzzy logic user and metadata interface apparatus</nc> according to <nc>claim</nc> 1 , further comprising: (d) <nc>a content layer</nc> operably connected to store <nc>a content resource and access module</nc>, and wherein said <nc>metadata layer maps</nc> said interactivity points to <nc>content resources</nc> stored in <nc>said content resource and access module</nc>.
|
8745725
|
13374479
|
1. <nc>A system</nc> comprising: <nc>a transfer</nc> determining <nc>module</nc> configured to determine that <nc>a computing device</nc>, <nc>that</nc> was presenting <nc>an item</nc>, has been transferred from <nc>a first user</nc> to <nc>a second user</nc>, <nc>the transfer</nc> determining <nc>module</nc> including at least: <nc>a visual cue</nc> detecting <nc>module</nc> configured to determine that <nc>the computing device</nc> has been transferred from <nc>the first user</nc> to <nc>the second user</nc> when <nc>the visual cue</nc> detecting <nc>module</nc> at least detects <nc>presence</nc> or <nc>absence</nc> of <nc>one or more visual cues</nc> in <nc>proximate vicinity</nc> of <nc>the computing device</nc> <nc>that</nc> when detected as occurring <nc>at least suggested transfer</nc> of <nc>the computing device</nc> between <nc>the first and second users</nc>, the visual cue detecting <nc>module</nc> including at least: <nc>a gesture</nc> detecting <nc>module</nc> configured to detect <nc>the presence</nc> or <nc>absence</nc> of <nc>the one or more visual cues</nc> in <nc>the proximate vicinity</nc> of <nc>the computing device</nc> when <nc>the gesture</nc> detecting <nc>module</nc> at least detects <nc>visually one or more gestures</nc> exhibited by <nc>the first user</nc> <nc>that</nc> when detected as occurring <nc>at least suggested transfer</nc> of <nc>the computing device</nc> from <nc>the first user</nc> to <nc>the second user</nc> at least in <nc>part</nc> by <nc>the first user</nc> moving <nc>the computing device</nc> at least in <nc>part</nc> with <nc>the one or more gestures</nc>; and a highlighted portion presenting <nc>module</nc> configured to present, via <nc>the computing device</nc>, <nc>one or more highlighted portions</nc> of <nc>the item</nc>, <nc>the highlighted portion</nc> presenting <nc>module</nc> being responsive at least in <nc>part</nc> to <nc>the transfer</nc> determining <nc>module</nc> configured to determine that <nc>a computing device</nc>, <nc>that</nc> was presenting <nc>an item</nc>, has been transferred from <nc>a first user</nc> to <nc>a second user</nc>, <nc>the highlighted portion</nc> presenting <nc>module</nc> being configured to present <nc>the one or more highlighted portions</nc> of <nc>the item</nc> responsive at least in <nc>part</nc> to <nc>the transfer</nc> determining <nc>module</nc> and to <nc>a highlighting selection</nc> detecting <nc>module</nc>, <nc>the highlighting selection detection module</nc> configured to detect, prior to <nc>the transfer</nc> of <nc>the computing device</nc> from <nc>the first user</nc> to <nc>the second user</nc>, that <nc>the first user</nc> has at least one of marked or tagged <nc>at least one or more parts</nc> of <nc>the one or more portions</nc> to select <nc>the one or more portions</nc> for highlighting.
| 21
|
21. <nc>The system</nc> of <nc>claim</nc> 1 , wherein said <nc>visual cue</nc> detecting <nc>module</nc> including <nc>at least a gesture</nc> detecting <nc>module</nc> configured to detect <nc>the presence</nc> or <nc>absence</nc> of <nc>the one or more visual cues</nc> in <nc>the proximate vicinity</nc> of <nc>the computing device</nc> when <nc>the gesture</nc> detecting <nc>module</nc> at least detects <nc>visually one or more gestures</nc> exhibited by <nc>the first user</nc> <nc>that</nc> when detected as occurring <nc>at least suggested transfer</nc> of <nc>the computing device</nc> from <nc>the first user</nc> to <nc>the second user</nc> at least in <nc>part</nc> by <nc>the first user</nc> moving <nc>the computing device</nc> at least in <nc>part</nc> with <nc>the one or more gestures</nc> comprises: <nc>a face</nc> detecting <nc>module</nc> configured to detect <nc>the presence</nc> or <nc>absence</nc> of <nc>the one or more visual cues</nc> in <nc>the proximate vicinity</nc> of <nc>the computing device</nc> when <nc>the face</nc> detecting <nc>module</nc> at least detects <nc>presence</nc> of <nc>a first face</nc> associated with <nc>the first user</nc> and <nc>a second face</nc> associated with <nc>the second user</nc> in <nc>the proximate vicinity</nc> of <nc>the computing device</nc>, <nc>the second face</nc> being detected as being closer to <nc>the computing device</nc> than <nc>the first face</nc>.
|
9704067
|
15178119
|
1. <nc>A method</nc> for automatically generating <nc>an image description</nc>, comprising following <nc>steps</nc> of: obtaining <nc>a first image data</nc>; analyzing <nc>a text file</nc> corresponding to <nc>the first image data</nc>; calculating <nc>occurrences</nc> and <nc>distribution ratios</nc> of <nc>terms</nc>, <nc>each</nc> of <nc>which</nc> contains <nc>a target word</nc>, from <nc>the text file</nc> so as to obtain <nc>a plurality</nc> of <nc>the terms</nc> having <nc>the distribution ratios</nc> greater than <nc>a threshold</nc>; comparing <nc>the distribution ratios</nc> of <nc>the terms</nc> so as to find out <nc>at least a key term</nc>; finding out one of <nc>a plurality</nc> of <nc>lexical chains</nc> containing <nc>the key term</nc> with <nc>the greatest distribution ratio</nc> so as to generate <nc>a narrative language chain</nc>; and setting <nc>the narrative language chain</nc> as <nc>a description</nc> of <nc>the first image data</nc>.
| 5
|
5. <nc>The method</nc> according to <nc>claim</nc> 1 , further comprising <nc>a step</nc> of: searching <nc>internet</nc> to find out <nc>a second image data</nc> correlating to <nc>the text file</nc>, <nc>the first image data</nc> or <nc>the narrative language chain</nc>.
|
7725346
|
11191776
|
1. <nc>A method</nc> of predicting <nc>an increase</nc> in <nc>sales</nc> from <nc>a plurality</nc> of <nc>online public discussions</nc>, comprising: receiving from <nc>a communications network</nc> <nc>a product information input</nc> for defining <nc>a product</nc> for <nc>which</nc> <nc>sales</nc> are predicted; receiving from <nc>the communications network</nc> <nc>a temporally defined input</nc> based on <nc>the number</nc> of <nc>times</nc> <nc>the product</nc> receives <nc>a mention</nc> in <nc>one or more sources</nc> of <nc>online chatter</nc>, wherein <nc>the temporally defined input</nc> comprises <nc>a mentions time series</nc> for <nc>the product</nc> derived from <nc>online chatter</nc>; generating by <nc>software</nc> installed on <nc>a computer host server</nc> <nc>a restriction</nc> <nc>that</nc> is automatically altered using <nc>a plurality</nc> of <nc>levels</nc> of <nc>disambiguation</nc> employing <nc>queries</nc> based on <nc>domain-specific keywords</nc> from <nc>the product information input</nc> to be applied to <nc>the temporally defined input</nc>; filtering with <nc>the software</nc> installed on <nc>the computer host</nc> server <nc>the temporally defined input</nc> with <nc>the restriction</nc>; generating by <nc>the software</nc> installed on <nc>the computer host server</nc>, <nc>a time-stamped temporally defined input</nc> from <nc>the filtered temporally defined input</nc>; generating by <nc>the software</nc> installed on <nc>the computer host server</nc> <nc>a signal</nc> quantifying <nc>the number</nc> of <nc>times</nc> <nc>the product</nc> is mentioned, wherein <nc>an amplitude</nc> of <nc>the signal</nc> is based on <nc>the number</nc> of <nc>times</nc> <nc>the product</nc> is mentioned between <nc>any two time-stamped temporally defined inputs</nc>; identifying <nc>a spike</nc> in <nc>the number</nc> of <nc>times</nc> <nc>the product</nc> is mentioned being present in <nc>the signal</nc>, wherein <nc>the spike</nc> is based on <nc>a rate</nc> of <nc>change</nc> in <nc>the amplitude</nc> exceeding <nc>a threshold value</nc> between <nc>the any two time-stamped temporally defined inputs</nc>, wherein <nc>the threshold value</nc> is based on <nc>a rate</nc> over <nc>time</nc> at <nc>which</nc> <nc>the product</nc> receives <nc>mentions</nc> from <nc>the plurality</nc> of <nc>online discussions</nc>; predicting with <nc>a processor</nc> <nc>a correlation value</nc> to create <nc>a comparison</nc> between <nc>the mentions time series</nc> and <nc>a sales rank time series</nc> of <nc>said product</nc>, wherein <nc>the sales rank time series</nc> represents <nc>measured sales</nc> of <nc>the product</nc>; where <nc>the correlation value</nc> is given by: r xy <nc></nc> ( <nc>k</nc> ) = <nc>c xy</nc> <nc></nc> ( k ) c xx <nc></nc> ( 0 ) · c yy <nc></nc> ( 0 ) where <nc>c xy</nc> is given by: <nc>c xy</nc> = 1 n <nc></nc> <nc>∑ t = 1 n - k </nc> ( x <nc>i</nc> - <nc>μ</nc> <nc></nc> ( x ) ) <nc> ( y i - μ </nc> ( <nc>y</nc> ) ) <nc></nc> k = 0 , … <nc></nc> , n - 1 , c xy = 1 n ∑ t = 1 - k n ( x i - <nc>μ</nc> ( x ) ) ( y i - μ ( <nc>y</nc> ) ) <nc></nc> k = - 1 , … , - ( n - 1 ) , and where <nc>μ</nc> is <nc>a sample mean</nc> and <nc>k</nc> is <nc>a lag value</nc> and where c xx (0) and <nc>c</nc> yy (0) are <nc>sample variances</nc> of <nc>the sales rank time series</nc> and <nc>the mentions time series</nc>, respectively; determining <nc>an optimum lag value</nc> for <nc>values</nc> of <nc>k</nc> where <nc>the correlation value</nc> is a maximum, wherein <nc>negative values</nc> for <nc>the optimum lag value</nc> are considered <nc>leading and non-negative values</nc> for <nc>the optimum lag value</nc> are considered trailing; predicting <nc>the increase</nc> in <nc>sales</nc> of <nc>the product</nc> from <nc>the identified spike</nc> of <nc>the signal</nc> if <nc>the correlation value</nc> is at least 0.5 and <nc>the optimum lag</nc> is leading; and adjusting <nc>the levels</nc> of <nc>disambiguation</nc> if either <nc>the optimum lag</nc> is not leading or <nc>the correlation value</nc> is less than 0.5.
| 2
|
2. <nc>The method</nc> of <nc>claim</nc> 1 , wherein predicting <nc>the increase</nc> in <nc>sales</nc> further comprises <nc>the spike</nc> exceeding <nc>historical averages</nc> by <nc>a predetermined threshold value</nc> of <nc>historical averages</nc>.
|
10083690
|
15458664
|
1. <nc>A method</nc> for operating <nc>a digital assistant</nc>, <nc>the method</nc> comprising: at <nc>an electronic device</nc> having <nc>one or more processors</nc> and <nc>memory</nc>: receiving <nc>user speech input</nc>; generating <nc>a textual representation</nc> of <nc>the user speech input</nc>; parsing <nc>the textual representation</nc> to determine <nc>a primary domain</nc> representing <nc>a user intent</nc> for <nc>the textual representation</nc>; identifying <nc>a first substring</nc> from <nc>the textual representation</nc> <nc>that</nc> corresponds to <nc>a first attribute</nc> of <nc>the primary domain</nc>; parsing <nc>the identified first substring</nc> to determine <nc>a secondary domain</nc> representing <nc>a user intent</nc> for <nc>the first sub string</nc>; performing <nc>a task flow</nc> comprising <nc>one or more tasks</nc> based on <nc>the primary domain</nc> and <nc>the secondary domain</nc>; and outputting <nc>a response</nc> in <nc>accordance</nc> with <nc>the performed task flow</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein parsing <nc>the identified first substring comprises</nc>: determining <nc>a confidence score</nc> for <nc>a plurality</nc> of <nc>interpretations</nc> of <nc>the first substring</nc>; and determining <nc>the secondary domain</nc> representing <nc>the user intent</nc> for <nc>the first substring</nc> based on <nc>an interpretation</nc> of <nc>the plurality</nc> of <nc>interpretations</nc> of <nc>the first substring</nc> having <nc>the highest confidence score</nc>.
|
9928231
|
15401446
|
1. <nc>A method</nc> for determining <nc>topics</nc> of <nc>short text messages</nc>, comprising: by <nc>a computer</nc>, obtaining <nc>distributed vector representations</nc> of <nc>words</nc> in <nc>a vocabulary</nc> identified in <nc>a corpus</nc> comprising <nc>a plurality</nc> of training <nc>short text messages</nc>, <nc>the distributed vector representations</nc> being obtained by processing <nc>windows</nc> of <nc>the corpus</nc> having <nc>a context window fixed length</nc>; by <nc>the computer</nc>, estimating <nc>a plurality</nc> of <nc>Gaussian components</nc> of <nc>a Gaussian mixture model</nc> of <nc>the corpus</nc> using <nc>the distributed vector representations</nc> and using <nc>bottleneck features</nc> obtained using <nc>neural networks</nc>, <nc>the Gaussian components</nc> representing <nc>corpus topics</nc>; by <nc>the computer</nc>, receiving <nc>a sample short text message</nc> comprising <nc>a subset</nc> of <nc>the words</nc> in <nc>the vocabulary</nc>; and by <nc>the computer</nc>, determining <nc>a topic</nc> of <nc>the sample short text message</nc> based on <nc>a posterior distribution</nc> over <nc>the corpus topics</nc> for <nc>the sample short text message</nc>, <nc>the posterior distribution</nc> obtained using <nc>the Gaussian mixture model</nc>.
| 2
|
2. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the training short text messages</nc> have <nc>a maximum message length</nc>, and <nc>the context window fixed length</nc> is greater than or equal to <nc>the maximum message length</nc>.
|
7933871
|
11966426
|
1. <nc>A processor-based system</nc> to manage <nc>template updates</nc> by: receiving <nc>an update request</nc> based in <nc>part</nc> on <nc>an update time</nc>, wherein <nc>the update request</nc> includes <nc>a request</nc> for <nc>one or more links</nc> associated with <nc>one or more document libraries</nc> having <nc>one or more templates</nc>, wherein <nc>the one or more links</nc> are targeted to <nc>an associated user</nc> and point to <nc>targets</nc> <nc>that</nc> include <nc>the one or more document libraries</nc> and <nc>associated templates</nc> used in <nc>part</nc> to maintain <nc>locally stored templates</nc> for <nc>the associated user</nc>; generating <nc>markup data</nc> for <nc>the one or more templates</nc> associated with <nc>the one or more links</nc>, wherein <nc>the markup data</nc> includes <nc>template parameters</nc>, <nc>a description</nc>, and <nc>other information</nc> associated with <nc>the one or more templates</nc> of <nc>the one or more document libraries</nc>; setting <nc>a template synchronization flag</nc> to identify <nc>a document library</nc> to use as <nc>part</nc> of <nc>an update process</nc> including associating <nc>a first group</nc> of <nc>users</nc> with <nc>a first set</nc> of <nc>document libraries</nc> and <nc>a second group</nc> of <nc>users</nc> with <nc>a second set</nc> of <nc>document libraries</nc>; determining whether to replace <nc>a local template</nc> with <nc>an associated template</nc> targeted by <nc>a link</nc> including comparing <nc>a first template parameter</nc> of <nc>the local template</nc> with <nc>a second parameter</nc> of <nc>the associated template</nc> of <nc>the document library</nc> including comparing <nc>local template attribute values</nc> of <nc>the local template</nc> with <nc>attribute values</nc> of <nc>the associated template</nc> of <nc>the document library</nc> associated with <nc>the link</nc>; maintaining <nc>the locally stored templates</nc> to correspond with <nc>new and updated templates</nc> associated with <nc>the one or more document libraries</nc> including determining <nc>which parts</nc> of <nc>a locally stored template</nc> require updating based in <nc>part</nc> on <nc>updated aspects</nc> of <nc>the associated template</nc>, including automatically updating <nc>the local template</nc> based in <nc>part</nc> on <nc>the comparison</nc> of <nc>the first and second template parameters</nc>; and storing <nc>template schemas</nc> and <nc>associated metadata</nc> locally with <nc>a client</nc> as <nc>part</nc> of <nc>the updating</nc>.
| 8
|
8. <nc>The system</nc> of <nc>claim</nc> 1 , further configured to manage <nc>template updates</nc> by using <nc>an updated template</nc> to create <nc>a work product</nc> including <nc>a site</nc> where <nc>users</nc> can create, organize, and share <nc>information</nc>, <nc>a site</nc> for <nc>document collaboration</nc>, and <nc>a site</nc> where <nc>users</nc> can post <nc>information</nc> and allow <nc>others</nc> to provide <nc>comments</nc>.
|
9122744
|
13271175
|
1. <nc>A method</nc> comprising: receiving from <nc>a client device</nc> and displaying on <nc>the client device</nc> <nc>a user inquiry</nc> associated with <nc>a user</nc>, <nc>the user inquiry</nc> having <nc>a linguistic pattern</nc> including <nc>a verb</nc>; generating <nc>a follow up question</nc> based on <nc>the user inquiry</nc>; displaying on <nc>the client device</nc> <nc>the follow up question</nc>; receiving from <nc>the client device</nc> and displaying on <nc>the client device</nc> <nc>a follow up answer</nc> from <nc>the user</nc>; determining <nc>a task</nc> to be performed at least in <nc>part</nc> by <nc>an intelligent assistant</nc> based at least in <nc>part</nc> on <nc>the follow</nc> up answer from <nc>the user</nc>, <nc>the task</nc> including at least one of buying <nc>an item</nc> or <nc>service</nc>, selling <nc>an item</nc> or <nc>service</nc>, <nc>publishing</nc>, sending <nc>a message</nc>, <nc>offering</nc>, comparing, making, automating, <nc>calling</nc>, <nc>setting</nc>, <nc>learning</nc>, <nc>saving</nc>, <nc>scheduling</nc>, subscribing, posting, starting, stopping, modifying, <nc>alerting</nc>, <nc>booking</nc>, or summarizing; causing <nc>the task</nc> to be performed at least in <nc>part</nc> by <nc>the intelligent assistant</nc>; and generating and displaying on <nc>the client device</nc> <nc>a response</nc> regarding <nc>the task</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 , <nc>wherein the user inquiry</nc>, <nc>the follow up question</nc>, <nc>the</nc> follow up answer, and <nc>the response</nc> are displayed via at least one of <nc>a browser plug-in</nc>, <nc>a browser input field</nc>, <nc>a mobile application</nc>, <nc>a web page</nc>, <nc>a browser toolbar</nc>, <nc>a computer application</nc>, <nc>a computer toolbar</nc>, <nc>a widget</nc> in <nc>a website</nc>, or <nc>an application programming interface</nc> (<nc>API</nc>).
|
10140976
|
14968439
|
1. <nc>A method</nc> for <nc>language processing</nc>, comprising: training <nc>one or more automatic speech recognition models</nc> using <nc>an automatic speech recognition dictionary and speech recognition training data</nc>; determining <nc>a set</nc> of N <nc>automatic speech recognition hypotheses</nc> <nc>that</nc> characterize <nc>a spoken input</nc>, based on <nc>the one or more automatic speech recognition models</nc>, using <nc>a processor</nc>; selecting <nc>a hypothesis</nc> from <nc>the set</nc> of N <nc>automatic speech recognition hypotheses</nc> using <nc>a discriminative language model</nc> and <nc>a first natural language processing dictionary</nc> <nc>that</nc> excludes <nc>words</nc> having <nc>little discriminatory value</nc> according to <nc>an error rate</nc> of <nc>only words</nc> other than <nc>words</nc> having <nc>little likely effect</nc> on <nc>the natural language outcome</nc> in <nc>each hypothesis</nc>; and performing <nc>natural language processing</nc> on <nc>the selected hypothesis</nc> using <nc>a second natural language processing dictionary</nc> <nc>that</nc> is different from <nc>the automatic speech recognition dictionary</nc> and <nc>the first natural language processing dictionary</nc>.
| 2
|
2. <nc>The method</nc> for <nc>language processing</nc> of <nc>claim</nc> 1 , further comprising determining <nc>the first natural language processing dictionary</nc> using <nc>natural language processing training data</nc> and <nc>the automatic speech recognition dictionary</nc> by trimming out <nc>words</nc> having <nc>little likely effect</nc> on <nc>the natural language outcome</nc> from <nc>the automatic speech recognition dictionary</nc>.
|
9898519
|
14028277
|
1. <nc>A method</nc>, including: receiving, for <nc>display</nc>, <nc>a set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> including <nc>information</nc> identifying <nc>multiple individuals</nc> or <nc>references</nc> to <nc>the set</nc> of <nc>prospect or contact objects</nc>; identifying <nc>a plurality</nc> of <nc>blank fields</nc> related to <nc>the multiple individuals</nc> identified by <nc>the received set</nc> of <nc>prospect or contact objects</nc>; upon receiving <nc>a selection</nc> from <nc>a user</nc> indicating <nc>a request</nc> to perform <nc>a social synchronization action</nc> to fill in <nc>the plurality</nc> of <nc>blank fields</nc> with <nc>multiple social identification handles</nc> of <nc>the multiple individuals</nc>: automatically spidering <nc>a plurality</nc> of <nc>social network platforms</nc> to obtain <nc>the multiple social identification handles</nc> of <nc>the multiple individuals</nc> identified by <nc>the received set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> or <nc>the references</nc> to <nc>the set</nc> of <nc>prospect or contact objects</nc>, wherein <nc>the social network platforms</nc> are independently operated; and updating <nc>the received set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> to include <nc>the obtained multiple social identification handles</nc> of <nc>the multiple individuals</nc>; and transmitting for display (<nc>i</nc>) data <nc>that</nc> includes <nc>at least some</nc> of <nc>the obtained multiple social identification handles</nc> or <nc>references</nc> to <nc>the obtained multiple social identification handles</nc> and <nc>(ii</nc>) <nc>an interface</nc> <nc>that</nc> shows that <nc>at least some</nc> of <nc>the blank fields</nc> of <nc>the received set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> have been filled in.
| 10
|
10. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: in <nc>response</nc> to obtaining <nc>the social identification handles</nc>, automatically retrieving <nc>conversation preferences</nc> and <nc>interests</nc> of <nc>the received set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> or <nc>the reference</nc> to <nc>the prospect</nc> or <nc>contact objects</nc>; and generating <nc>an interface</nc> <nc>that</nc> highlights <nc>the conversation preferences</nc> and <nc>interests</nc> of <nc>the received set</nc> of <nc>prospect</nc> or <nc>contact objects</nc> or <nc>the reference</nc> to <nc>the prospect</nc> or <nc>contact objects</nc>.
|
9780805
|
14520485
|
1. <nc>A computer program product</nc> for <nc>predicate application</nc> using <nc>partial compression dictionary match</nc>, <nc>the computer program product</nc> comprising <nc>a computer readable storage medium</nc> having <nc>program instructions</nc> embodied therewith, wherein <nc>the computer readable storage medium</nc> is not <nc>a transitory signal</nc> per se, <nc>the program instructions</nc> being executable by <nc>a processor</nc> to cause <nc>the processor</nc> to perform <nc>a method</nc> comprising: developing <nc>a search strategy</nc> for <nc>each predicate</nc> to be applied to <nc>compressed data</nc>, <nc>the search strategy</nc> comprising: searching <nc>entries</nc> of <nc>a compression dictionary</nc> for <nc>an entire predicate value</nc> to identify <nc>each matching symbol</nc> and noting <nc>an exact match</nc>, and a subsumed match of <nc>a compression symbol</nc> in <nc>a search list</nc>; searching <nc>the entries</nc> of <nc>the compression dictionary</nc> to identify <nc>entries</nc> in <nc>which</nc> <nc>a predicate prefix</nc> forms <nc>a dictionary entry suffix</nc> and noting <nc>each matching compression symbol</nc> in <nc>the search list</nc>; searching <nc>the entries</nc> of <nc>the compression dictionary</nc> to identify <nc>entries</nc> in <nc>which</nc> <nc>a predicate</nc> <nc>suffix</nc> forms <nc>a dictionary entry prefix</nc> and noting <nc>each matching compression symbol</nc> in <nc>the search list</nc>; searching <nc>the entries</nc> of <nc>the compression dictionary</nc> to identify <nc>entries</nc> <nc>which</nc> form <nc>a mid-predicate match</nc> and noting <nc>each matching compression symbol</nc> in <nc>the search list</nc>; and adding <nc>one or more uncompressed symbols</nc> for <nc>predicates</nc> having <nc>no match</nc> in <nc>the compression dictionary</nc>, and for <nc>predicates</nc> having <nc>only one or more subsumed dictionary matches</nc>; searching <nc>the compressed data</nc> to locate <nc>the compression symbols</nc> identified in <nc>the search list</nc>; and in <nc>response</nc> to locating <nc>a compression symbol</nc> from <nc>the search list</nc> in <nc>the compressed data</nc>, decompressing <nc>a respective row</nc> and applying <nc>the predicate</nc>, and returning <nc>a respective row</nc> <nc>that</nc> matches <nc>the predicate</nc> to one of: <nc>a database engine</nc> and <nc>an application</nc>.
| 7
|
7. <nc>The computer program product</nc> of <nc>claim</nc> 1 , wherein <nc>the compressed data</nc> is <nc>multi-level compressed data</nc>.
|
7747616
|
11477623
|
1. <nc>A file search system</nc> for searching for <nc>a file</nc> stored in <nc>a storage device</nc>, comprising: <nc>reception</nc> means for receiving <nc>a search request</nc> made by <nc>a user</nc>; <nc>search</nc> means for obtaining <nc>search result files</nc> by searching <nc>the storage device</nc> on <nc>the basis</nc> of <nc>notification</nc> from the reception means; monitor <nc>means</nc> for monitoring <nc>a file operation</nc> by <nc>the user</nc>; <nc>creation</nc> means for creating <nc>a reference relation</nc> of <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the file operation</nc> by <nc>the user</nc> monitored by <nc>the monitor means</nc>, <nc>the reference relation</nc> being defined as <nc>a relation</nc> between <nc>a reference file</nc> and <nc>a reference source file</nc> in <nc>which</nc> <nc>information</nc> included in <nc>the reference file</nc> is introduced into <nc>the reference source file</nc> by <nc>the file operation</nc> of <nc>the user</nc>, wherein <nc>the reference file</nc> is <nc>a file</nc> to be referenced and <nc>the reference source file</nc> is <nc>a file</nc> referencing <nc>the reference file</nc>, and <nc>the introduction</nc> of <nc>information</nc> from <nc>the reference file</nc> to <nc>the reference source file</nc> is performed by copying <nc>information</nc> from <nc>the reference file</nc> and pasting <nc>the copied information</nc> into <nc>the reference source file</nc>; <nc>storage</nc> means for storing <nc>the reference relation</nc>; <nc>importance calculation</nc> means for calculating <nc>an importance</nc> of <nc>the reference file</nc> on <nc>the basis</nc> of <nc>one or more items</nc> selected from among <nc>reference relation items</nc> <nc>each</nc> obtained by monitoring <nc>a file operation</nc> of <nc>the user</nc> regarding <nc>introduction</nc> of <nc>information</nc> from <nc>the reference file</nc> to <nc>the reference source file</nc>; and <nc>search result sort</nc> means for arranging <nc>the search result files</nc> obtained by the search means on <nc>the basis</nc> of <nc>the importance</nc> of <nc>the reference file</nc>; and a control section for displaying <nc>a screen</nc> for allowing <nc>the user</nc> to select <nc>arrangement</nc> of <nc>the search result files</nc> obtained by the search means on <nc>the basis</nc> of <nc>the importance</nc> or <nc>a condition</nc> other than <nc>the importance</nc>, <nc>the control section</nc> judging whether or not <nc>the user</nc> selects <nc>the arrangement</nc> of <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the importance</nc>; wherein <nc>the search result sort</nc> means arranges <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the importance</nc> when <nc>the arrangement</nc> of <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the importance</nc> is selected, and arranges <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the condition</nc> other than <nc>the importance</nc> when <nc>the arrangement</nc> of <nc>the search result files</nc> on <nc>the basis</nc> of <nc>the importance</nc> is not selected.
| 4
|
4. <nc>The file search system</nc> according to <nc>claim</nc> 1 , further comprising: <nc>weight setting rule</nc> obtaining <nc>means</nc> for obtaining <nc>a weight setting rule</nc> for setting <nc>a weight</nc> of <nc>the reference relation</nc>; <nc>identification</nc> means for identifying <nc>the user</nc>; and <nc>recording</nc> <nc>means</nc> for recording <nc>the weight setting rule</nc> in <nc>association</nc> with <nc>each user</nc>, wherein <nc>the importance calculation</nc> means calculates <nc>the importance</nc> on <nc>the basis</nc> of <nc>the weight setting rule</nc> associated with <nc>the user</nc> <nc>who</nc> has made <nc>the search request</nc>.
|
9317551
|
13834671
|
1. <nc>A method</nc> comprising: receiving <nc>information</nc> associated with <nc>a plurality</nc> of <nc>technical computing environment</nc> (<nc>TCE) models</nc>, <nc>the</nc> receiving <nc>the information</nc> being performed by <nc>a device</nc>; executing <nc>the plurality</nc> of <nc>TCE models</nc> to generate <nc>execution information</nc> associated with <nc>the plurality</nc> of <nc>TCE models</nc>, <nc>the</nc> executing <nc>the plurality</nc> of <nc>TCE models</nc> being performed by <nc>the device</nc>; storing <nc>the plurality</nc> of <nc>TCE models</nc> and <nc>the execution information</nc> for <nc>association</nc> with <nc>a TCE-based search engine</nc>, <nc>the</nc> storing <nc>the plurality</nc> of <nc>TCE models</nc> being performed by <nc>the device</nc>; providing, for <nc>display</nc>, <nc>a user interface</nc> associated with <nc>the TCE-based search engine</nc>, the providing, for <nc>display</nc>, <nc>the user interface</nc> being performed by <nc>the device</nc>; receiving <nc>a query</nc> via <nc>the user interface</nc>, <nc>the</nc> receiving <nc>the query</nc> being performed by <nc>the device</nc>; dividing <nc>the query</nc> into <nc>one or more query elements</nc>, the dividing <nc>the query</nc> being performed by <nc>the device</nc>; identifying <nc>a group</nc> of <nc>query elements</nc> of <nc>the one or more query elements</nc>, the identifying being performed by <nc>the device</nc>; processing <nc>the group</nc> of <nc>the query elements</nc> based on at least one of <nc>query content</nc> or <nc>information</nc> requested by <nc>the query</nc>, the processing <nc>the group</nc> of <nc>the query elements</nc> being performed by <nc>the device</nc>; transforming <nc>the query</nc> into <nc>another query</nc> based on <nc>the processed group</nc> of <nc>the query elements</nc>, <nc>the other query</nc> being different than <nc>the query</nc>, and <nc>the</nc> transforming <nc>the query</nc> being performed by <nc>the device</nc>; determining <nc>a respective input type</nc> for <nc>each processed query element</nc> of <nc>the processed group</nc> of <nc>the query elements</nc>, <nc>each respective input type</nc> corresponding to <nc>a respective TCE model</nc> of <nc>the plurality</nc> of <nc>TCE models</nc>, and the determining <nc>the respective input type</nc> being performed by <nc>the device</nc>; selecting <nc>at least one TCE model</nc>, of <nc>the plurality</nc> of <nc>TCE models</nc>, based on <nc>the determined respective input type</nc>, the selecting <nc>the at least one TCE model</nc> being performed by <nc>the device</nc>; and providing <nc>the other query</nc> to <nc>the selected at least one TCE model</nc> for <nc>further processing</nc>, <nc>the</nc> providing <nc>the other query</nc> being performed by <nc>the device</nc>.
| 6
|
6. <nc>The method</nc> of <nc>claim</nc> 1 , where <nc>the other query</nc> includes at least one of: <nc>model input data</nc>, a model configuration parameter, <nc>a model configuration setting</nc>, <nc>tool data</nc>, <nc>a tool configuration parameter</nc>, or <nc>personal information</nc> about <nc>a user</nc> associated with <nc>the query</nc>.
|
9501471
|
13909386
|
1. <nc>A method</nc> for generating <nc>a context</nc> for translating <nc>strings</nc> for <nc>a graphical user interface</nc>, <nc>the method</nc> comprising: receiving <nc>a string</nc> to be translated and <nc>associated source code</nc>, <nc>the string</nc> identified by <nc>a unique key</nc> within <nc>the associated source code</nc>, <nc>the unique key</nc> used to identify <nc>a location</nc> of <nc>the string</nc>; identifying, by <nc>one or more computer processors</nc>, <nc>a first logical section</nc> of <nc>the associated source code</nc> corresponding to <nc>a unique key</nc> of <nc>the string</nc>; identifying, by <nc>the one or more computer processors</nc>, <nc>one or more graphical user interface components</nc> within <nc>the identified logical section</nc> of <nc>the associated source code</nc>; and creating <nc>a mockup image</nc> presenting <nc>the one or more graphical user interface components</nc> and <nc>the string</nc>.
| 2
|
2. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising presenting <nc>the created mockup image</nc> to <nc>a translator</nc>.
|
9100319
|
14496355
|
1. <nc>A method</nc> comprising: receiving, by <nc>a network interface</nc> of <nc>a network device</nc>, <nc>a packet stream</nc>; identifying <nc>one or more packets</nc> within <nc>the packet stream</nc> <nc>that</nc> satisfy <nc>one or more conditions</nc> of <nc>a plurality</nc> of <nc>predefined conditions</nc> specified by <nc>a rule</nc> by pre-matching, by <nc>an acceleration device</nc> of <nc>the network device</nc>, <nc>the packet stream</nc> with <nc>the plurality</nc> of <nc>predefined conditions</nc>; identifying, by <nc>the acceleration device</nc>, for <nc>a full-match processing stage</nc> <nc>a candidate packet</nc> of <nc>the one or more packets</nc> <nc>that</nc> satisfies <nc>all</nc> of <nc>the plurality</nc> of <nc>predefined conditions</nc> by correlating <nc>the one or more satisfied conditions</nc> of <nc>the candidate packet</nc>; generating, by <nc>the acceleration device</nc>, for <nc>use</nc> by <nc>the full-match processing stage</nc> <nc>a matching token</nc> and <nc>a corresponding location</nc> of <nc>the matching token</nc> within <nc>the candidate packet</nc> for <nc>each</nc> of <nc>the plurality</nc> of <nc>predefined conditions</nc>; and determining, by <nc>a context-aware pattern matching and parsing (CPMP) processor</nc> of <nc>the acceleration device</nc>, whether <nc>the candidate packet</nc> meets <nc>the rule</nc> by performing <nc>the full-match processing stage</nc> including fetching and executing <nc>special purpose CPMP instructions</nc> to perform <nc>context-aware pattern matching processing</nc> on <nc>the packet</nc>, wherein <nc>the context-aware pattern matching processing</nc> includes one or more of <nc>string matching</nc>, <nc>regular expression matching</nc> and <nc>packet field</nc> <nc>value matching</nc> based on, for <nc>each</nc> of <nc>the plurality</nc> of <nc>predefined conditions</nc>, <nc>corresponding contextual information</nc> provided by <nc>the rule</nc>, <nc>the matching token</nc> and <nc>the corresponding location</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising performing <nc>reordering</nc> of <nc>packets</nc> in <nc>the packet stream</nc>.
|
9740879
|
14526823
|
1. <nc>A computer-implemented method</nc> for using <nc>searchable encryption</nc> to query <nc>a database</nc> storing <nc>encrypted data</nc>, <nc>the method</nc> being executed using <nc>one or more processors</nc> and comprising: executing, by <nc>the one or more processors</nc>, at least a portion of <nc>a searchable encryption scheme</nc> <nc>that</nc> provides <nc>an asymptotically optimal, sub-linear search time</nc> and comprises <nc>probabilistic operations</nc> to generate <nc>a set</nc> of <nc>search indices</nc> and <nc>a search token</nc>; receiving, by <nc>the one or more processors</nc>, the set of <nc>search indices</nc>; receiving, by <nc>the one or more processors</nc>, the search token, and in <nc>response</nc>: searching, by <nc>the one or more processors</nc>, at least, <nc>one search index</nc> of <nc>the set</nc> of <nc>search indices</nc> based on <nc>the search token</nc>, and <nc>determining</nc>, by <nc>the one or more processors</nc>, that <nc>the at least one search index</nc> corresponds to <nc>an initial setup protocol</nc>, wherein <nc>the at least one search index</nc> comprises <nc>an empty table</nc> <nc>that</nc> is absent <nc>an entry</nc> and is absent <nc>a search history</nc> of <nc>previously submitted search tokens</nc> corresponding to <nc>the search token</nc>, and in <nc>response</nc>, receiving <nc>one or more identifiers</nc>, <nc>each identifier</nc> being associated with <nc>a respective ciphertext</nc> comprising <nc>an access pattern</nc> <nc>that</nc> is determined to be responsive to <nc>the search token</nc>, and updating <nc>the at least one search index</nc> to comprise <nc>the entry</nc> based on <nc>the search token</nc> and <nc>the one or more identifiers</nc> and to comprise <nc>the search history</nc> of the search token by using <nc>an inverted index</nc> for <nc>the search token</nc>; and transmitting, by <nc>the one or more processors</nc>, search results, <nc>the search results</nc> comprising <nc>the one or more ciphertexts</nc> <nc>that</nc> are determined to be responsive to <nc>the search token</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the set</nc> of <nc>search indices</nc> are provided based on <nc>a set</nc> of <nc>keys</nc> stored at <nc>a client-side computing device</nc> and are received from <nc>the client-side computing device</nc>, <nc>at least one key</nc> in <nc>the set</nc> of <nc>keys</nc> being used to provide <nc>the encrypted data</nc> stored in <nc>the database</nc>.
|
9390709
|
14032906
|
1. <nc>A semiconductor integrated circuit device</nc> comprising: <nc>a signal processing unit</nc> extracts frequency components of <nc>an inputted voice signal</nc>, and generates <nc>a feature pattern</nc> representing <nc>a state</nc> of <nc>distribution</nc> of <nc>the frequency components</nc> of <nc>the voice signal</nc>; <nc>a voice recognition database storage unit</nc> <nc>which</nc> stores <nc>a plurality</nc> of <nc>voice recognition databases</nc> <nc>each</nc> including <nc>a standard pattern</nc> representing <nc>a state</nc> of <nc>distribution</nc> of <nc>frequency components</nc> of <nc>plural phonemes</nc> used in <nc>a predetermined language</nc>, <nc>each</nc> of <nc>the plurality</nc> of <nc>voice recognition databases</nc> having been generated based on <nc>voice signals</nc> of <nc>a different group</nc> of <nc>speakers</nc>, <nc>each different group</nc> of <nc>speakers</nc> having <nc>a different age</nc> and/or <nc>gender</nc> than <nc>other ones</nc> of <nc>the groups</nc> of <nc>speakers</nc>; <nc>a conversion list storage unit</nc> <nc>which</nc> stores <nc>a conversion list</nc> expressed by <nc>character data</nc> and including <nc>plural words</nc> or <nc>sentences</nc> to be <nc>conversion candidates</nc>, <nc>the plural words</nc> or <nc>candidates</nc> being expected <nc>responses</nc> to <nc>a question</nc> or <nc>message</nc>; <nc>a standard pattern extraction unit</nc> <nc>which</nc> extracts <nc>the standard pattern</nc> corresponding to <nc>the character data</nc> representing <nc>the first syllable</nc> of <nc>each word</nc> or <nc>sentence</nc> included in <nc>the conversion list</nc>, from <nc>the voice recognition database</nc>; and <nc>a matching detection unit</nc> <nc>which</nc> receives <nc>an input</nc> of <nc>age</nc> and/or <nc>gender</nc> of <nc>a user</nc>, and selects a corresponding one of <nc>the plurality</nc> of <nc>voice recognition databases</nc> based on <nc>the input</nc>, and <nc>which</nc> compares <nc>the feature pattern</nc> generated from <nc>the first syllable</nc> of <nc>the voice signal</nc> with <nc>the standard pattern</nc> extracted by <nc>the standard pattern extraction unit</nc>, thus detects <nc>the matching</nc> of <nc>the syllable</nc>, and <nc>outputs</nc> <nc>information</nc> specifying <nc>a word</nc> or <nc>sentence</nc> <nc>that</nc> has <nc>the matching-detected syllable</nc> as <nc>the first syllable</nc> thereof.
| 3
|
3. <nc>The semiconductor integrated circuit device</nc> according to <nc>claim</nc> 1 , wherein if <nc>the conversion list</nc> includes <nc>plural words</nc> or <nc>sentences</nc> having <nc>the matching-detected syllable</nc> as <nc>the first syllables</nc> thereof, <nc>the matching detection unit</nc> expands <nc>a range</nc> of <nc>syllables</nc> to be detected for matching.
|
8615664
|
12231302
|
1. <nc>A system</nc>, comprising: <nc>an inference data acquisition module</nc> configured to acquire <nc>inference data</nc> <nc>that</nc> indicate <nc>an inferred mental state</nc> of <nc>an authoring user</nc> in <nc>connection</nc> with <nc>a particular item</nc> of <nc>an electronic message</nc>, <nc>the inference data</nc> derived based, at least in <nc>part</nc>, on <nc>at least one physical characteristic</nc> of <nc>the authoring user</nc>; <nc>one or more sensors</nc> configured to sense <nc>the at least one physical characteristic</nc> of <nc>the authoring user</nc> in <nc>connection</nc> with <nc>the particular item</nc> of <nc>the electronic message</nc>; <nc>a source identity acquisition module</nc> configured to acquire <nc>source identity data</nc> providing <nc>at least one identity</nc> of <nc>one or more sources</nc> <nc>that</nc> provide <nc>a basis</nc>, at least in <nc>part</nc>, for <nc>the inference data</nc>, <nc>the one or more sources</nc> including <nc>at least the one or more sensors</nc>; <nc>an inference data association module</nc> configured to associate <nc>the inference data</nc> with <nc>the particular item</nc>, <nc>the inference data association module</nc> including <nc>at least an inference data inclusion module</nc> configured to include <nc>the inference data</nc> into <nc>the electronic message</nc>; and <nc>a source identity association module</nc> configured to associate <nc>the source identity data</nc> with <nc>the particular item</nc>, the source identity association module including <nc>at least a source identity inclusion module</nc> configured to include into <nc>the electronic message</nc> <nc>one or more identities</nc> of <nc>the one or more sensors</nc>, <nc>the one or more sensors</nc> having been used to derive, at least in <nc>part</nc>, <nc>the inference data</nc> acquired by <nc>the inference data acquisition module</nc>; and wherein <nc>the electronic message</nc> thereby includes <nc>at least a data pair</nc> <nc>that</nc> includes <nc>at least the inference data</nc> <nc>that</nc> indicate <nc>the inferred mental state</nc> of <nc>the authoring user</nc> in <nc>connection</nc> with <nc>the particular item</nc> and <nc>the one or more identities</nc> of <nc>the one or more sensors</nc> used to derive, at least in <nc>part</nc>, <nc>the inference data</nc>.
| 40
|
40. <nc>The system</nc> of <nc>claim</nc> 1 , wherein said <nc>inference data acquisition module</nc> configured to acquire <nc>inference data</nc> <nc>that</nc> indicate <nc>an inferred mental state</nc> of <nc>an authoring user</nc> in <nc>connection</nc> with <nc>a particular item</nc> of <nc>an electronic message</nc>, <nc>the inference data</nc> derived based, at least in <nc>part</nc>, on <nc>at least one physical characteristic</nc> of <nc>the authoring user</nc> comprises: <nc>an inference data acquisition module</nc> configured to acquire <nc>an indication</nc> of <nc>an action</nc> executed, at least in <nc>part</nc>, by <nc>the authoring user</nc> in <nc>connection</nc> with <nc>the particular item</nc>.
|
7858868
|
11631944
|
1. <nc>A method</nc> for operating <nc>a digital processing apparatus</nc> including <nc>a processor</nc> for classifying <nc>music</nc>, comprising: (a) providing <nc>music classification data</nc> <nc>that</nc> are descriptive for <nc>a discrete and finite set</nc> of <nc>a finite number</nc> of <nc>music classes</nc>; (b) providing <nc>an unclassified piece</nc> of <nc>music</nc> to be classified; (c) deriving, in <nc>the processor</nc>, for <nc>each</nc> of <nc>said music classes</nc> of said <nc>set</nc> of <nc>music classes</nc>, <nc>a respective Gish distance value</nc> with <nc>respect</nc> to <nc>said unclassified piece</nc> of <nc>music</nc> to be classified to obtain <nc>a discrete and finite set</nc> of <nc>a finite number</nc> of <nc>Gish distance values</nc>, wherein said <nc>finite set</nc> of <nc>a finite number</nc> of <nc>Gish distance values</nc> is descriptive for <nc>a relation</nc> of <nc>said unclassified piece</nc> of <nc>music</nc> to be classified with <nc>respect</nc> to said <nc>discrete and finite set</nc> of <nc>a finite number</nc> of <nc>music classes</nc>.
| 8
|
8. <nc>The method</nc> according to <nc>claim</nc> 1 , wherein said providing (<nc>a</nc>) said <nc>music classification data</nc> includes, receiving or generating, in <nc>the processor</nc>, said <nc>music classification data</nc> or <nc>a part</nc> thereof.
|
8224656
|
12049243
|
1. <nc>A method</nc> for providing <nc>speech disambiguation</nc> on <nc>a mobile device</nc>, comprising: transmitting <nc>audio</nc> for <nc>speech recognition processing</nc>; receiving <nc>results</nc> <nc>representing alternates</nc> identified to match <nc>the transmitted audio</nc>; caching <nc>the received results</nc>, <nc>the cache</nc> utilizing <nc>a least recently used (LRU) algorithm</nc> to handle <nc>cache overflow</nc>; searching for <nc>content</nc> associated with <nc>the received results</nc>; displaying <nc>the alternates</nc> in <nc>a disambiguation dialog screen</nc> for making <nc>corrections</nc> to <nc>the alternates</nc>; making <nc>corrections</nc> to <nc>the alternates</nc> using <nc>the disambiguation dialog screen</nc> until <nc>a correct result</nc> is displayed, wherein <nc>the correction</nc> to <nc>the alternates</nc> using <nc>the disambiguation dialog screen</nc> and <nc>the receiving</nc> of <nc>the results</nc> representing <nc>the alternates</nc> are performed in <nc>parallel</nc> to <nc>the search</nc> for <nc>the content</nc> associated with <nc>the received results</nc>; selecting <nc>the correct result</nc>; upon selecting <nc>the correct result</nc>, mapping <nc>the correct result</nc> to one of <nc>the received results</nc> in <nc>the cache</nc>, <nc>the cache</nc> deleting <nc>unselected results</nc> prior to utilizing <nc>the LRU</nc>; and immediately displaying <nc>a matching associated search result</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 further comprising <nc>displaying content</nc> associated with <nc>the selected correct result</nc> and received in <nc>parallel</nc> with <nc>the receiving</nc> of <nc>the results</nc> representing <nc>alternates</nc> identified to match <nc>the transmitted audio</nc>.
|
7599920
|
11549075
|
1. <nc>A computer-implemented method</nc> of <nc>indexing documents</nc> in <nc>websites</nc>, <nc>the method</nc> comprising: on <nc>a server system</nc> having <nc>one or more processors</nc> and memory storing <nc>programs</nc> to be executed by <nc>the one or more processors</nc>: for <nc>each website</nc> of <nc>a multiplicity</nc> of <nc>websites</nc>, <nc>each website</nc> having <nc>a corresponding current crawl rate limit</nc>: crawling <nc>the respective website</nc>, in <nc>accordance</nc> with <nc>the current crawl rate limit</nc> corresponding to <nc>the respective website</nc>, to download <nc>documents</nc> from <nc>the respective website</nc> for <nc>inclusion</nc> in <nc>a database</nc>; storing <nc>crawl data</nc> associated with <nc>the crawling</nc> of <nc>the respective website</nc>; providing, for <nc>display</nc>, <nc>a crawl rate control mechanism</nc> to <nc>a respective owner</nc> of <nc>the respective website</nc>, including providing for <nc>display</nc> to <nc>the respective owner</nc> <nc>at least a portion</nc> of <nc>the crawl data</nc>, and enabling <nc>selection</nc> of <nc>a new crawl rate limit</nc> corresponding to <nc>the respective website</nc> by <nc>the respective owner</nc>; comparing <nc>a maximum crawl rate</nc> for <nc>the respective website</nc> over <nc>a defined period</nc> of <nc>time</nc> with <nc>the current crawl rate limit</nc> for crawling <nc>the respective website</nc> to determine if <nc>the current crawl rate limit</nc> is <nc>a limiting factor</nc> in crawling <nc>the respective website</nc>; and in <nc>response</nc> to <nc>a request</nc> to increase <nc>a current crawl rate</nc> for crawling <nc>the respective website</nc>, increasing <nc>the current crawl rate limit</nc> only when <nc>the current crawl rate limit</nc> is <nc>a limiting factor</nc> in crawling <nc>the respective website</nc>.
| 4
|
4. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , further comprising: when <nc>the current crawl rate limit</nc> is not <nc>a limiting factor</nc> in crawling <nc>the respective website</nc>, informing <nc>the respective owner</nc> that <nc>a faster crawl rate</nc> may not be selected.
|
8589449
|
12272444
|
1. <nc>A network storage system</nc> comprising: <nc>data storage</nc> <nc>that</nc> <nc>stores file data</nc> and <nc>metadata</nc> associated with <nc>the file data</nc>, <nc>the file data</nc> and <nc>metadata</nc> received from <nc>at least first and second computers</nc> <nc>each</nc> running <nc>a different operating system</nc>, with <nc>each different operating system</nc> defining <nc>a different file system</nc>, <nc>wherein each</nc> of <nc>the different file systems</nc> for <nc>the first and second computers stores file data</nc> and <nc>metadata</nc> associated with <nc>the file data</nc> in <nc>corresponding first and second different formats</nc>; <nc>one or more application program interfaces</nc> (<nc>APIs</nc>) in <nc>communication</nc> with <nc>the data storage</nc> <nc>that</nc> define <nc>operations</nc> for reading and writing <nc>the file data</nc> and <nc>metadata</nc> in <nc>the first and second different formats</nc>; and <nc>a metadata handler</nc> having <nc>a library</nc> of <nc>functions</nc> <nc>that</nc> handle <nc>at least the metadata</nc> in <nc>the first and second formats</nc>, <nc>the library</nc> of <nc>functions</nc> comprising <nc>a metadata object extraction function</nc> <nc>that</nc> accesses <nc>a metadata object</nc> <nc>having populated information</nc> corresponding to <nc>all metadata fields</nc> used by <nc>the first and second formats</nc>, <nc>the populated information</nc> being in <nc>a third format</nc> different from <nc>the first and second formats</nc>, extracts <nc>the populated information</nc> from <nc>the metadata object</nc>, and cooperates with at least one of <nc>the one or more APIs</nc> to generate <nc>at least a portion</nc> of <nc>the metadata</nc> in at least one of <nc>the first and second formats</nc> from <nc>the extracted populated information</nc>.
| 4
|
4. <nc>The system</nc> of <nc>claim</nc> 1 , further comprising <nc>at least one client application</nc> <nc>that</nc> invokes <nc>the metadata handler library</nc>.
|
8166003
|
11418746
|
1. <nc>A computer-implemented method</nc> of providing <nc>an audience-appropriate version</nc> of <nc>the document</nc>, <nc>the computer-implemented method</nc> being arranged to cause <nc>a computer</nc> to perform <nc>actions</nc>, comprising: receiving <nc>a request</nc> for <nc>a document</nc> at <nc>a document access point</nc> associated with <nc>said document</nc>, said <nc>document</nc> comprising <nc>a plurality</nc> of <nc>document versions</nc>, wherein said <nc>request</nc> is associated with <nc>file access authorizations</nc>, and wherein said <nc>plurality</nc> of <nc>document versions</nc> are associated with <nc>file access permissions</nc>; determining <nc>which</nc> of <nc>the plurality</nc> of <nc>document versions</nc> of <nc>the document</nc> has <nc>a majority</nc> of <nc>previous document accesses</nc>; pre-fetching <nc>the determined document version</nc> for <nc>the document</nc> <nc>that</nc> has <nc>the majority</nc> of <nc>previous document accesses</nc>; identifying based on <nc>a match</nc> between <nc>a file access authorization</nc> associated with <nc>the request</nc> and <nc>a file access permission</nc> corresponding to <nc>the identified document version</nc> one of <nc>said plurality</nc> of <nc>document versions</nc>, wherein <nc>the file access permissions</nc> of said identified of <nc>document versions</nc> correspond to said <nc>file access authorizations</nc>, such that <nc>the file access permissions</nc> include <nc>a security level value</nc> indicating <nc>a security level</nc> of <nc>the document version</nc> and <nc>an audience field</nc> indicating <nc>an authorized audience</nc>, wherein <nc>each</nc> of <nc>the plurality</nc> of <nc>the document versions</nc> is uniquely identified by <nc>a combination</nc> of <nc>a name</nc> of <nc>the documents</nc> and <nc>the security level</nc> of <nc>the document version</nc>, wherein <nc>the security level</nc> indicates <nc>an audience</nc> allowed <nc>access</nc> to <nc>the document</nc>; wherein <nc>each</nc> of <nc>the plurality</nc> of <nc>document versions</nc> include <nc>a name field</nc>, <nc>a level field</nc> and <nc>the audience field</nc>; and retrieving said one of <nc>said plurality</nc> of <nc>document versions</nc>.
| 4
|
4. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , wherein said <nc>file access permissions</nc> comprise one of <nc>a plurality</nc> of <nc>access levels</nc>.
|
7987443
|
11590705
|
1. <nc>A computer-based user interface</nc> comprising: <nc>a dialog control element</nc> <nc>that</nc> executes on <nc>a computer</nc> for <nc>usage</nc> in creating <nc>a dialog</nc>, configured to receive and act upon <nc>a data model</nc> containing <nc>recursive data structures</nc> comprising <nc>at least one object</nc> defined by <nc>at least one property</nc> and modifiable by <nc>at least one control</nc>, <nc>the dialog control element</nc> further configured to associate <nc>object properties</nc> with <nc>controls</nc> recursively whereby <nc>properties</nc> contain <nc>objects</nc> <nc>which</nc> further contain <nc>properties</nc> and responsive to <nc>a user request</nc> to modify <nc>objects</nc> by modifying <nc>user interface controls</nc> and modifying <nc>portions</nc> of <nc>the recursive data structure</nc> to reflect <nc>object modifications</nc>, wherein <nc>the dialog control element</nc> performs <nc>recursive operations</nc> by retrieving <nc>a template</nc> and repeats <nc>a dialog interaction</nc> for <nc>object properties</nc> contained in <nc>the at least one object</nc> and creates <nc>at least one control</nc> recursively corresponding to <nc>the object properties</nc> and <nc>the template</nc>.
| 4
|
4. <nc>The computer-based user interface</nc> according <nc>to claim</nc> 1 wherein: <nc>the dialog control element</nc> is configured as <nc>a runtime library</nc> for <nc>linkage</nc> into <nc>an application</nc>, creating <nc>a dialog specification</nc>.
|
9263037
|
12761207
|
1. <nc>An interactive manual system</nc> of <nc>a device</nc>, comprising: <nc>a model package database</nc> storing: <nc>a set</nc> of at least one of <nc>phrases</nc> and <nc>sentences</nc>; <nc>an object file</nc> <nc>that</nc> includes <nc>data objects</nc> <nc>that</nc> include <nc>information</nc> regarding <nc>an appearance</nc> of, and <nc>spatial characteristics</nc> of, <nc>a plurality</nc> of <nc>structures</nc> of <nc>the device</nc> and <nc>that</nc> include <nc>data</nc> representing <nc>visual displays</nc> <nc>that</nc> are associated with <nc>the structures</nc> of <nc>the device</nc>; and <nc>a grammar table</nc> <nc>that</nc> identifies <nc>associations</nc> between (a) <nc>the information</nc> and <nc>data</nc> and <nc>(b</nc>) <nc>respective ones</nc> of <nc>the set</nc> of the at least one of <nc>phrases</nc> and <nc>sentences</nc>; and <nc>a processor</nc> configured to execute: <nc>a speech engine module</nc> for converting <nc>an utterance</nc> into <nc>a word sequence</nc>, determining <nc>the word sequence's meaning</nc>, <nc>which</nc> is associated with one or more of <nc>the set</nc> of the at least one of <nc>phrases</nc> and <nc>sentences</nc>, and identifying <nc>a grammar category</nc> to <nc>which</nc> the word sequence as <nc>a whole</nc> conforms; and a dialog manager module for extracting <nc>a subset</nc> of <nc>the information</nc> regarding <nc>the appearance</nc> of, and spatial characteristics of, <nc>the one or more structures</nc> of <nc>the device</nc>; and an output arrangement configured to output <nc>what</nc> has been extracted; wherein <nc>the extraction</nc> is based on <nc>the identified grammar category</nc>, the one or more of <nc>the set</nc> of the at least one of <nc>phrases</nc> and <nc>sentences</nc> to <nc>which</nc> <nc>the determined meaning</nc> of <nc>the word sequence</nc> corresponds, and <nc>the associations</nc> of <nc>the grammar table</nc>, such that <nc>the extraction</nc> includes collecting from <nc>the data</nc> objects <nc>a plurality</nc> of <nc>data objects</nc> <nc>that concern</nc> a particular one of <nc>the structures</nc> of <nc>the device</nc> associated by <nc>the grammar table</nc> with the one or more of <nc>the set</nc> of the at least one of <nc>phrases</nc> and <nc>sentences</nc> <nc>that</nc> are associated with <nc>the determined meaning</nc> of <nc>the word sequence</nc>, <nc>different ones</nc> of the data objects <nc>that concern</nc> <nc>the particular structure</nc> being included as <nc>part</nc> of <nc>the collected plurality</nc> of <nc>data objects</nc> depending on <nc>the category</nc> to <nc>which</nc> <nc>the processor</nc> has identified <nc>the word sequence</nc> as <nc>a whole</nc> conforms.
| 4
|
4. <nc>The interactive manual system</nc> of <nc>claim</nc> 1 , wherein <nc>the output arrangement</nc> includes <nc>a speech synthesis arrangement</nc>.
|
9053102
|
13755987
|
1. <nc>A method</nc> for deriving and utilizing <nc>a context object</nc> to generate <nc>a synthetic context-based object</nc>, <nc>the method</nc> comprising: deriving, by <nc>one or more processors</nc>, <nc>a context object</nc> for <nc>a non-contextual data object</nc>, wherein <nc>the non-contextual data</nc> object ambiguously relates to <nc>multiple subject-matters</nc>, wherein <nc>the context object</nc> provides <nc>a context</nc> <nc>that</nc> identifies <nc>a specific subject-matter</nc>, from <nc>multiple subject-matters</nc>, of <nc>the non-contextual data object</nc>, and wherein <nc>the context object</nc> is derived by contextually searching and analyzing <nc>a document</nc>, <nc>which</nc> contains <nc>multiple instances</nc> of <nc>the non-contextual data object</nc>, to derive <nc>the context object</nc>; establishing <nc>a minimum validity threshold</nc> for <nc>the context object</nc>, wherein <nc>the minimum validity threshold</nc> defines <nc>a probability</nc> that <nc>a derived context object</nc> accurately describes <nc>the context</nc> of <nc>the non-contextual data object</nc>; expanding <nc>a range</nc> of <nc>a search area</nc> of <nc>the document</nc> until <nc>the minimum validity threshold</nc> is reached; associating, by <nc>one or more processors</nc>, the non-contextual data object with <nc>the context object</nc> to define <nc>a synthetic context-based object</nc>; associating, by <nc>one or more processors</nc>, <nc>the synthetic context-based object</nc> with <nc>at least one specific data store</nc>, wherein said <nc>at least one specific data store</nc> comprises <nc>data</nc> <nc>that</nc> is associated with <nc>data</nc> contained in <nc>the non-contextual data object</nc> and <nc>the context object</nc>; constructing, by <nc>one or more processors</nc>, a dimensionally constrained hierarchical synthetic context-based object library for <nc>multiple synthetic context-based objects</nc>, wherein <nc>synthetic context-based objects</nc> within <nc>a same dimension</nc> of <nc>the dimensionally constrained hierarchical synthetic context-based object library share data</nc> from <nc>a same non-contextual data object</nc>, and wherein <nc>synthetic context-based objects</nc> within <nc>the same dimension</nc> of <nc>the dimensionally constrained hierarchical synthetic context-based object library</nc> contain <nc>disparate data</nc> from <nc>different context objects</nc>; receiving, from <nc>a requester</nc>, <nc>a request</nc> for <nc>at least one data store</nc> <nc>that</nc> is associated with <nc>synthetic context-based objects</nc> within <nc>the same dimension</nc> of <nc>the dimensionally constrained hierarchical synthetic context-based object library</nc>; and returning, to <nc>the requester</nc>, said <nc>at least one specific data store</nc> <nc>that</nc> is associated with <nc>synthetic context-based objects</nc> within <nc>the same dimension</nc> of <nc>the dimensionally constrained hierarchical synthetic context-based object library</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the specific subject-matter</nc> for <nc>a particular data store</nc> in <nc>the data structure</nc> is exclusive to only said <nc>particular data store</nc>.
|
8533199
|
13031615
|
1. <nc>A computer-implemented method</nc> for creating <nc>an intelligent bookmark</nc> to <nc>a document</nc>, comprising: displaying <nc>the document</nc> comprising <nc>at least an address</nc>, <nc>a title</nc>, and <nc>a body</nc>; receiving <nc>a user selection</nc> of <nc>content</nc> from <nc>a portion</nc> of <nc>the body</nc> of <nc>the document</nc> and <nc>user</nc> supplemented <nc>information</nc>; generating <nc>the intelligent bookmark</nc> by retrieving <nc>the address</nc> of <nc>the document</nc>, and by automatically extracting <nc>identifier information</nc> from <nc>the body</nc> of <nc>the document</nc> within <nc>the user selection</nc>; and storing <nc>the address</nc> in <nc>association</nc> with <nc>the identifier information</nc> and <nc>the user</nc> supplemented <nc>information</nc>.
| 11
|
11. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the intelligent bookmark</nc> includes <nc>an image</nc> and further comprising: receiving <nc>a mark-up</nc> of <nc>the intelligent bookmark</nc>; and generating <nc>a layer</nc> to display <nc>the mark-up</nc>.
|
9575980
|
14828685
|
1. <nc>A method</nc> of operating <nc>an information management system</nc>, comprising: obtaining <nc>a first collection</nc> of <nc>source files</nc>; for <nc>each source file</nc> in <nc>the first collection</nc>: parsing, by <nc>a processor</nc> of <nc>a computing device</nc>, the respective source file to extract <nc>one or more tags</nc>, comparing, by <nc>the processor</nc>, the one or more tags to <nc>tags</nc> in <nc>at least one dictionary</nc> to determine <nc>one or more matching tags</nc>, wherein <nc>the at least one dictionary</nc> comprises <nc>a hierarchical listing</nc> of <nc>tags</nc>, and associating, by <nc>the processor</nc>, with <nc>the respective source file</nc>, <nc>the one or more matching tags</nc>; generating, by <nc>the processor</nc>, a first virtual relational network comprising <nc>the source files</nc> in <nc>the first collection</nc>, wherein <nc>the first virtual relational network</nc> comprises: <nc>one or more nodes</nc>, wherein <nc>each node</nc> of <nc>the one or more nodes</nc> represents <nc>a particular matching tag</nc> of <nc>the one or more matching tags</nc> associated with <nc>a particular source file</nc> of <nc>the source files</nc> in <nc>the first collection</nc>, and <nc>one or more links</nc>, <nc>wherein each link</nc> of <nc>the one or more links</nc> represents <nc>a connection</nc> between <nc>a pair</nc> of <nc>nodes</nc>, wherein <nc>each node</nc> of <nc>the pair</nc> of <nc>nodes</nc> is associated with <nc>a same tag</nc>; comparing, by <nc>the processor</nc>, <nc>the first virtual relational network</nc> to <nc>a second virtual relational network</nc> to identify at least one of (<nc>a) nodes</nc> common to <nc>the first and second virtual relational networks</nc>, and <nc>(b) links</nc> common to <nc>the first and second virtual relational networks</nc>, wherein <nc>the second virtual relational network</nc> is created from <nc>a second collection</nc> of <nc>source files</nc> different from <nc>the first collection</nc> of <nc>source files</nc>, and <nc>the second virtual relational network</nc> is created using <nc>one or more dictionaries</nc> of <nc>the at least one dictionary</nc>; and displaying <nc>a graphical representation</nc> of <nc>at least part</nc> of <nc>the first and second virtual relational networks</nc>.
| 3
|
3. <nc>The method</nc> of <nc>claim</nc> 1 wherein <nc>tags</nc> comprise <nc>representations</nc> of <nc>embedded objects</nc>.
|
7984008
|
11955671
|
1. <nc>A non-transitory computer readable storage medium</nc>, comprising <nc>executable instructions</nc> to: identify <nc>sub-string repetition</nc> in <nc>a pass phrase</nc>; generate <nc>a sub-string index</nc>; determine <nc>the number</nc> of <nc>bits</nc> required to represent <nc>each character</nc> in <nc>the pass phrase</nc>; assign <nc>an entropy value</nc> to <nc>each character</nc> in <nc>the pass phrase</nc> in <nc>accordance</nc> with <nc>a probability function</nc> <nc>that</nc> assigns <nc>highest probabilities</nc> at <nc>boundaries</nc> of <nc>an interval</nc> for <nc>the entropy</nc> of <nc>a character</nc>; and compute <nc>a total entropy value</nc> for <nc>the pass phrase</nc>.
| 2
|
2. <nc>The computer readable storage medium</nc> of <nc>claim</nc> 1 wherein <nc>the executable instructions</nc> to identify include <nc>executable instructions</nc> to identify <nc>a sub</nc><nc>-</nc><nc>string</nc> as <nc>one or more characters</nc> <nc>that</nc> occurs at least twice in <nc>the pass phrase</nc>.
|
9177320
|
11295103
|
1. <nc>A method</nc> to generate <nc>a data file</nc> related to <nc>a particular entity</nc>, <nc>the method</nc> comprising: performing <nc>a first search</nc>, by <nc>a processor</nc>, with <nc>use</nc> of <nc>a first search term</nc> related to <nc>the particular entity</nc>, wherein <nc>performance</nc> of <nc>the first search</nc> includes sending <nc>the first search term</nc> over <nc>a network</nc>; receiving, by <nc>the processor</nc>, first results from <nc>the first search</nc>, wherein <nc>the first results</nc> relate to <nc>the particular entity</nc> and wherein <nc>the first results</nc> include <nc>first data</nc> in <nc>a first structure</nc>; <nc>parsing</nc>, by <nc>the processor</nc>, <nc>the first data</nc> from <nc>the first structure</nc> to produce <nc>first pieces</nc> of <nc>unstructured data</nc>; receiving, by <nc>the processor</nc>, <nc>second results</nc> from <nc>the first search</nc>, wherein <nc>the second results</nc> relate to <nc>the particular entity</nc> and wherein <nc>the second results</nc> include <nc>second data</nc> in <nc>a second structure</nc>, wherein <nc>the second structure</nc> is different from <nc>the first structure</nc>; <nc>parsing</nc>, by <nc>the processor</nc>, <nc>the second data</nc> from <nc>the second structure</nc> to produce <nc>second pieces</nc> of <nc>unstructured data</nc>; identifying <nc>matching pieces</nc> of <nc>data</nc> from among <nc>the first and second pieces</nc> of <nc>unstructured data</nc>; combining <nc>the matching pieces</nc> of <nc>data</nc> to form <nc>combined data</nc> in <nc>a third structure</nc> different from <nc>the first structure</nc> and <nc>the second structure</nc>, wherein <nc>the third structure</nc> relates to <nc>the particular entity</nc> as opposed to <nc>other entities</nc>, and wherein <nc>the combined data</nc> in <nc>the third structure</nc> relates to <nc>the particular entity</nc>; storing <nc>the combined data</nc> in <nc>the third structure</nc> in <nc>a memory</nc>; receiving <nc>a second search query</nc> related to <nc>the entity</nc>; performing <nc>the second search</nc> of <nc>the combined data</nc> in <nc>the third structure</nc>, by <nc>the processor</nc>, with <nc>use</nc> of <nc>the second search query</nc> related to <nc>the entity</nc>; producing <nc>a data file</nc> by <nc>the processor</nc> based on <nc>the results</nc> of <nc>the second search</nc>, wherein <nc>the data file</nc> includes <nc>at least some</nc> of <nc>the first and second pieces</nc> of <nc>unstructured data</nc>; and storing <nc>the data file</nc> in <nc>the memory</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the third structure</nc> includes <nc>a database</nc> related to <nc>the particular entity</nc>.
|
10140266
|
14145076
|
1. <nc>A method</nc> comprising: in <nc>a computer system</nc> comprising <nc>at least a processor</nc> and <nc>a memory</nc>, generating a XSL (<nc>Extensible Stylesheet Language</nc>) Transformations (<nc>XSLT</nc>) and Extensible Markup Language <nc>(XML</nc>) Path Language (XPath) execution tree for <nc>a source XSLT stylesheet</nc>; determining <nc>one or more optimizations</nc> for <nc>the XSLT and XPath execution tree</nc>, the determining <nc>one or more optimizations</nc> for <nc>the XSLT and XPath execution tree</nc> comprising <nc>profiling</nc> with <nc>sample data</nc>, identifying <nc>one or more hot-spot execution instruction nodes</nc>, and identifying <nc>one or more patterns</nc> for <nc>optimization</nc> for <nc>the one or more identified hot-spot execution instruction nodes</nc>, the identifying <nc>one or more hot-spot execution instruction nodes</nc> comprising, for <nc>each instruction</nc> in <nc>the execution tree</nc>, analyzing <nc>time</nc> and count values compared to <nc>threshold values</nc>, and selecting <nc>instructions</nc> for <nc>further analysis</nc> if <nc>predetermined conditions</nc> are fulfilled; applying <nc>the one or more optimizations</nc> to <nc>the XSLT and XPath execution tree</nc>; verifying <nc>the one or more optimizations</nc> in <nc>the XSLT and XPath execution tree</nc>; making <nc>the verified one or more optimizations</nc> persistent in <nc>an optimized source</nc> <nc>XSLT stylesheet</nc>; and transforming <nc>one or more source XML documents</nc> into <nc>one or more result documents</nc> using <nc>the optimized source XSLT stylesheet</nc>.
| 2
|
2. <nc>A method</nc> of <nc>claim</nc> 1 wherein <nc>profiling</nc> with <nc>sample data</nc> comprises: transforming <nc>data</nc> <nc>a non-empty set</nc> of <nc>input files</nc> with <nc>data</nc> considered <nc>representative</nc> of <nc>actual data</nc> with <nc>hit-count profiling</nc> enabled in <nc>multiple passes</nc>; and summarizing <nc>the profile data</nc> of <nc>each</nc> of <nc>the multiple passes</nc>.
|
9817813
|
14150628
|
1. <nc>A method</nc> comprising: receiving, on <nc>a computer system</nc> comprising <nc>a processor</nc> and <nc>memory</nc> storing <nc>instructions</nc>, <nc>a supplied phrase</nc>, <nc>the supplied phrase</nc> comprising <nc>one or more terms</nc>, <nc>the supplied phrase</nc> being associated with <nc>a category</nc> of <nc>a plurality</nc> of <nc>categories</nc>, <nc>each category</nc> being associated with <nc>a different topic</nc> and <nc>a plurality</nc> of <nc>phrases</nc> <nc>each</nc> having <nc>a meaning</nc> semantically related to <nc>the different topic</nc>, <nc>the plurality</nc> of <nc>categories</nc> being used by <nc>an analytics system</nc> to perform <nc>classifications</nc>; examining <nc>one or more terms</nc> included in <nc>the supplied phrase</nc>; based on examining <nc>the one or more terms</nc>, determining that <nc>a first term</nc> of <nc>the supplied phrase</nc> corresponds to <nc>a semantic group</nc>; identifying <nc>a second term</nc> <nc>that</nc> is included in <nc>the semantic group</nc>; generating, using <nc>the supplied phrase</nc> and <nc>the second term</nc>, <nc>a suggested phrase</nc> having <nc>a similar meaning</nc> to <nc>the supplied phrase</nc>, <nc>the suggested phrase</nc> and <nc>the supplied phrase</nc> being semantically related to <nc>the different topic</nc> associated with <nc>the category</nc>; and adding <nc>the suggested phrase</nc> to <nc>the category</nc> <nc>that</nc> includes <nc>the supplied phrase</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the semantic group</nc> comprises <nc>a plurality</nc> of <nc>terms</nc>, and wherein <nc>the first term</nc> is replaced with <nc>the second term</nc> from <nc>the plurality</nc> of <nc>terms</nc> of <nc>the semantic group</nc>, <nc>the second term</nc> being different from <nc>the first term</nc>.
|
9946712
|
14897510
|
1. <nc>A computer-implemented method</nc>, comprising: outputting, at <nc>a computing device</nc> including <nc>one or more processors</nc>, <nc>a media stream</nc> <nc>that</nc> includes at least one of <nc>audio data and video data</nc>, <nc>the media stream</nc> being fully defined between <nc>a start time</nc> and <nc>an end time</nc>; during <nc>the outputting</nc> of <nc>the media stream</nc>: <nc>(i</nc>) receiving, at <nc>the computing device</nc>, <nc>a user input</nc> (a) identifying <nc>a temporal point</nc> within <nc>the media stream</nc> and (b) indicating <nc>a request</nc> for translating <nc>a portion</nc> of <nc>the media stream</nc> from <nc>a source language</nc> to <nc>a target language</nc>, (<nc>ii</nc>) in <nc>response</nc> to receiving <nc>the user input</nc>, <nc>determining</nc>, at <nc>the computing device</nc>, <nc>a text</nc> by performing one of <nc>(a) speech recognition</nc> on <nc>a sub</nc><nc>-</nc><nc>period</nc> of <nc>the audio data</nc> as specified by <nc>the identified temporal point</nc> within <nc>the media stream</nc> and <nc>(b</nc>) <nc>optical character recognition</nc> on <nc>an image</nc> from <nc>the video data</nc> as specified by <nc>the identified temporal point</nc>, and <nc>(iii) transmitting</nc>, from <nc>the computing device</nc>, <nc>the text</nc> to <nc>a translation server</nc> via <nc>a network</nc>; receiving, at <nc>the computing device</nc>, <nc>a translated text</nc> from <nc>the translation server</nc> via <nc>the network</nc>, <nc>the translated text</nc> having been translated from <nc>the source language</nc> to <nc>the target language</nc> by <nc>the translation server</nc>; and in <nc>response</nc> to <nc>completion</nc> of <nc>the media stream</nc>, displaying, at <nc>a display</nc> of <nc>the computing device</nc>, the translated text at <nc>the end time</nc> of <nc>the media stream</nc>.
| 2
|
2. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , wherein the identified audio data sub<nc>-</nc>period is <nc>a predetermined period</nc> of <nc>the audio data</nc> prior to <nc>the temporal point</nc> until <nc>the temporal point</nc>.
|
8392825
|
12850350
|
1. <nc>A method</nc> comprising: dividing <nc>a first version</nc> of <nc>a document</nc> into <nc>one or more sections</nc>; generating <nc>a condensed version</nc> of <nc>the document</nc> <nc>that</nc> includes <nc>a linkcorresponding</nc> to at least one of <nc>the one or more sections</nc>; transmitting <nc>the condensed version</nc> of <nc>the document</nc> to <nc>a mobile device</nc> for <nc>display</nc>; receiving <nc>a modified version</nc> of <nc>the document</nc> from <nc>the mobile device</nc>, <nc>the modified version</nc> including <nc>one or more edits</nc> to one or more of <nc>the sections</nc>; and re-aggregating <nc>the modified one or more sections</nc> with <nc>unmodified sections</nc> to form <nc>a revised document</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the one or more sections</nc> correspond to <nc>one or more paragraphs</nc> in <nc>the first version</nc> of <nc>the document</nc>.
|
10095486
|
13581182
|
1. <nc>A method</nc> of developing <nc>a software application</nc> in <nc>which</nc> <nc>a check-in terminal</nc> communicates with <nc>two or more airline systems</nc> and <nc>services</nc>, <nc>the method</nc> comprising: providing <nc>a common code base</nc> to <nc>a computer system</nc>, wherein <nc>the common code base</nc> is incompatible with <nc>the two or more airline systems</nc> and <nc>services</nc>; describing <nc>data</nc> and <nc>messages</nc> used by <nc>the application</nc> using <nc>a declarative data description language</nc>; defining <nc>a library</nc> of <nc>system components</nc>, stored in <nc>a store</nc> of <nc>the computer system</nc>, including: a) <nc>a terminal abstraction layer</nc> <nc>that</nc> allows <nc>interactions</nc> between <nc>the check-in terminal</nc> and <nc>the application</nc> to control <nc>the check-in terminal</nc>; and <nc>b) an airline systems and services abstraction layer</nc> <nc>that</nc> allows <nc>interactions</nc> between <nc>the two or more airline systems</nc> and <nc>services</nc> and <nc>the application</nc>, wherein <nc>the two or more airline systems</nc> and <nc>services</nc> comprise <nc>check-in systems</nc>; providing <nc>a graphical (GUI) tool</nc> to model <nc>the workflow</nc> of <nc>the application</nc>, <nc>the workflow</nc> including <nc>screens</nc> and <nc>services</nc> described declaratively by <nc>a declarative data description language</nc>, <nc>the graphical tool</nc> being for <nc>display</nc> on <nc>a display device</nc> of <nc>the computer system</nc>; and assembling <nc>the application</nc> using <nc>the graphical tool</nc>, <nc>declarative rules</nc>, and <nc>customizations</nc> of <nc>system components</nc> selected from <nc>the library</nc>.
| 4
|
4. <nc>A method</nc> according to <nc>claim</nc> 1 , wherein <nc>the graphical tool</nc> provides <nc>access</nc> to <nc>the two or more airline systems</nc>.
|
7836428
|
11082204
|
1. <nc>A computer readable storage device</nc> having <nc>computer readable program instructions</nc> embodied <nc>thereon</nc> for programming <nc>a processor</nc>, said <nc>instructions</nc> consisting of: <nc>a computer software application</nc> manifesting <nc>an entirely declarative system</nc> for creating <nc>a software program</nc> comprising <nc>properties</nc> for declaratively establishing <nc>relationships</nc> between <nc>data elements</nc>, said <nc>relationships</nc> comprising <nc>descriptions</nc> of <nc>logic</nc> and <nc>data</nc>, wherein said <nc>entirely declarative system</nc> comprises <nc>one or more declarative lattice structures</nc>, <nc>each respective lattice</nc> having <nc>a plurality</nc> of <nc>configurable constructs</nc> comprising: (<nc>i</nc>) <nc>one or more declarative attributes</nc><nc>, each attribute</nc> configurable to select <nc>an internal behavior</nc> of <nc>said respective lattice</nc>; <nc>(ii</nc>) <nc>one or more declarative data access sites</nc>, <nc>each site</nc> configurable to define <nc>respective data access points</nc> for accessing <nc>lattices</nc> external to said <nc>respective lattice</nc>, and wherein said <nc>attributes and data access sites</nc> configure <nc>each</nc> said <nc>respective lattice</nc> to complete <nc>a singular computing task</nc>, and <nc>each respective lattice</nc> is configured to instantiate and execute, either alone or in <nc>relation</nc> to <nc>other lattices</nc> of <nc>said declarative system</nc>.
| 38
|
38. <nc>The computer readable storage device</nc> of <nc>claim</nc> 1 , wherein said <nc>one or more declarative lattice structures</nc> comprises <nc>a sorting structure</nc> configurable to sort <nc>a collection</nc> of <nc>data</nc>, wherein one or more of said <nc>attributes</nc> determine <nc>a sort behavior</nc> from <nc>a plurality</nc> of <nc>available sort behaviors</nc>, and wherein one of <nc>said sites</nc> is configurable to <nc>access</nc> said <nc>collection</nc> of <nc>data</nc>.
|
9189746
|
13349306
|
1. <nc>A method</nc> performed by <nc>one or more processors</nc> configured with <nc>computer-executable instructions</nc>, <nc>the method</nc> comprising: receiving <nc>an account</nc> associated with <nc>information</nc> including <nc>an email address</nc>; extracting <nc>one or more features</nc> from <nc>the information</nc> associated with <nc>the account</nc>, wherein at least one of <nc>the one or more features</nc> is based on <nc>memorability</nc> of <nc>the email address</nc>, <nc>the memorability</nc> relating to <nc>a pattern</nc> of <nc>symmetry</nc>, <nc>anti</nc><nc>-</nc><nc>symmetry</nc>, or uniformly distanced <nc>characters</nc> in <nc>the email address</nc>; and determining <nc>a trust level</nc> of <nc>the account</nc> at least partly based on <nc>the extracted features</nc>.
| 2
|
2. <nc>The method</nc> as recited in <nc>claim</nc> 1 , further comprising: determining that <nc>the account</nc> is benign if <nc>the determined trust level</nc> is higher than <nc>a first preset threshold</nc>; and/or determining that <nc>the account</nc> is malicious if <nc>the determined trust level</nc> is lower than <nc>the first preset threshold</nc> or <nc>a second preset threshold</nc> <nc>that</nc> is different from <nc>the first preset threshold</nc>.
|
9715635
|
14241908
|
1. <nc>An apparatus</nc> for <nc>document identification</nc>, having: <nc>a capture device</nc> for capturing <nc>a document feature</nc> of <nc>a document</nc>; <nc>a processor</nc> designed to perform <nc>document identification</nc> locally using <nc>the document feature</nc> if <nc>a processing criterion</nc> for <nc>the local performance</nc> of <nc>document identification</nc> by <nc>the apparatus</nc> for <nc>document identification</nc> is satisfied; and <nc>a transmitter</nc> designed to send <nc>a data record</nc> <nc>that</nc> is dependent on <nc>the document feature</nc> via <nc>a communication network</nc> to <nc>a communication network address</nc> if <nc>the processing criterion</nc> for <nc>the local performance</nc> of <nc>document identification</nc> by <nc>the apparatus</nc> for <nc>document identification</nc> is not satisfied; wherein <nc>the processing criterion</nc> is satisfied if <nc>the available processing resources</nc> of <nc>the apparatus</nc> are sufficient for performing <nc>document identification</nc>, or if <nc>the size</nc> of <nc>the document feature</nc> is below <nc>a prescribed threshold value</nc>, or if <nc>a data transmission speed</nc> of <nc>the communication network</nc> is below <nc>a threshold value</nc>, and wherein <nc>the processing criterion</nc> is not satisfied, if <nc>the available processing resources</nc> of <nc>the apparatus</nc> are not sufficient for performing <nc>document identification</nc>, or if <nc>the size</nc> of <nc>the document feature</nc> exceeds <nc>a prescribed threshold value</nc>, or if <nc>a connection speed</nc> via <nc>the communication network</nc> is below <nc>a threshold value</nc>.
| 7
|
7. <nc>The apparatus</nc> as claimed in <nc>claim</nc> 1 , wherein <nc>the processor</nc> is designed to produce <nc>the data record</nc> <nc>that</nc> is dependent on <nc>the document feature</nc> by <nc>anonymization</nc> of <nc>the document feature</nc>, <nc>which</nc> prevents <nc>reconstruction</nc> of <nc>the document feature</nc>, or by <nc>transformation</nc> of <nc>the document feature</nc>, or by <nc>segmentation</nc> of <nc>a transform</nc> of <nc>the document feature</nc>, or by speeded up <nc>robust feature detection</nc> (<nc>SURF</nc>), or by <nc>production</nc> of <nc>a hologram</nc> from <nc>the document feature</nc>, or by <nc>production</nc> of <nc>a histogram</nc> on <nc>the basis</nc> of <nc>the document feature</nc>, or by <nc>optical character recognition</nc> (<nc>OCR</nc>), or by <nc>capture</nc> of <nc>microprint</nc>.
|
8941733
|
11995130
|
<nc>1. Video retrieval system</nc> connectable to <nc>a plurality</nc> of <nc>surveillance cameras</nc> for monitoring <nc>a surroundings</nc>, <nc>the video retrieval system</nc> comprising: displaying <nc>means</nc> for displaying <nc>a graphical overall description</nc> of <nc>the surroundings</nc> under <nc>surveillance</nc>, whereby <nc>the graphical overall description</nc> comprises <nc>a plurality</nc> of <nc>monitored areas</nc> <nc>that</nc> are allocated to <nc>different surveillance cameras</nc> and whereby <nc>the graphical overall description</nc> is displayed in <nc>a single perspective view</nc>, and querying <nc>means</nc> comprising <nc>a graphical user interface</nc> configured to enable <nc>a user</nc> to formulate <nc>a trajectory-specific search query</nc> graphically and/or interactively on <nc>the graphical overall description</nc>, <nc>the search query</nc> adapted to retrieve <nc>data</nc> about <nc>a moving object</nc> in <nc>the surroundings</nc> with <nc>a matching trajectory</nc>, and wherein <nc>the graphical user interface</nc> enables <nc>users</nc> to draw <nc>any</nc> of <nc>the group</nc> consisting of <nc>lines</nc>, <nc>curves</nc>, <nc>circles</nc>, <nc>hand-drawn lines</nc>, <nc>a spline</nc> generated by <nc>a plurality</nc> of <nc>single interpolation points</nc> of <nc>the graphical overall description</nc> and <nc>any combination</nc> thereof, thereby formulating <nc>the search query</nc>, wherein <nc>the video retrieval system</nc> is characterized by merging <nc>means</nc> for merging <nc>object-related information</nc> of <nc>the video data</nc> of <nc>the plurality</nc> of <nc>cameras</nc> into <nc>object-related meta-data</nc>, and wherein <nc>the video retrieval system</nc> is characterized in that the merging means is realized to merge <nc>the video data</nc> into <nc>the overall description</nc> and to extract <nc>object-related meta-data</nc> from <nc>the overall description</nc> and/or to extract <nc>object-related information</nc> from <nc>the video data</nc> of <nc>the plurality</nc> of <nc>cameras</nc> and to link and/or merge <nc>the object-related information</nc> to generate <nc>the object-related meta-data</nc>.
| 6
|
<nc>6. Video retrieval system</nc> according to <nc>claim</nc> 1 , characterized in that <nc>the querying means</nc> is adapted for performing <nc>the trajectory-specific search query</nc> on <nc>the meta-data</nc>.
|
8996629
|
13225209
|
1. <nc>A computer-implemented method</nc> for generating <nc>a stream</nc> of <nc>content</nc> for <nc>each</nc> of <nc>a plurality</nc> of <nc>channels</nc>, <nc>the method</nc> comprising: generating, with <nc>one or more processors</nc>, <nc>a model</nc> for <nc>a user</nc> comprising <nc>an interest</nc> of <nc>the user</nc> and <nc>prior interaction</nc> of <nc>the user</nc> with <nc>heterogeneous data sources</nc>; computing, with <nc>the one or more processors</nc>, <nc>an interestingness score</nc> for <nc>each content item</nc> received from <nc>the heterogeneous data sources</nc> by summing <nc>properties</nc> of <nc>each content item</nc> over <nc>single-attribute properties</nc> using <nc>the model</nc> and based upon <nc>interestingness</nc> of <nc>each content item</nc> to <nc>the user</nc> and <nc>an extent</nc> to <nc>which</nc> <nc>the content item's popularity</nc> has increased within <nc>a geographic area</nc> associated with <nc>the user</nc>; categorizing, with <nc>the one or more processors</nc>, <nc>content items</nc> received from <nc>the heterogeneous data sources</nc> by annotating <nc>each content item</nc> with <nc>a dynamic feature</nc> including <nc>the interestingness score</nc>; identifying, with <nc>the one or more processors</nc>, <nc>a first channel category</nc> for <nc>the user</nc> based on <nc>a historical trend</nc> and <nc>the prior interaction</nc> of <nc>the user</nc> with <nc>the heterogeneous data sources</nc>, <nc>the historical trend</nc> including <nc>a change</nc> in <nc>a number</nc> of <nc>content items</nc> categorized under <nc>the first channel category</nc>; receiving <nc>an input</nc> through <nc>a user interface</nc> specifying <nc>a second channel category</nc>; querying <nc>the content items</nc> based on <nc>the first channel category</nc>, <nc>the second channel category</nc> and <nc>at least one channel attribute</nc>; in <nc>response</nc> to <nc>the query</nc>, receiving <nc>candidate content items</nc> <nc>that</nc> include <nc>the first channel category</nc>, <nc>the second channel category</nc> and <nc>the at least one channel attribute</nc> and comparing <nc>the interestingness score</nc> for <nc>each candidate content item</nc> with <nc>a threshold</nc> for <nc>the first channel category</nc> and <nc>the second channel category</nc> to determine <nc>the candidate content items</nc> <nc>that</nc> have <nc>an interestingness score</nc> <nc>that</nc> exceeds <nc>the threshold</nc>; and generating <nc>the stream</nc> of <nc>content</nc> from <nc>the candidate content items</nc> <nc>that</nc> have <nc>an interestingness score</nc> <nc>that</nc> exceeds <nc>the threshold</nc>.
| 6
|
6. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising receiving <nc>a request</nc> from <nc>the user</nc> to subscribe to <nc>an existing channel</nc>.
|
9892362
|
14546417
|
1. <nc>A method</nc>, in <nc>a data processing system</nc> comprising <nc>a processor</nc> and <nc>a memory</nc>, for identifying and evaluating <nc>hypothetical ontological links</nc> in <nc>an ontology</nc> and outputting <nc>information</nc>, <nc>the method</nc> comprising: receiving, by <nc>the data processing system</nc>, <nc>an ontology</nc> comprising <nc>a plurality</nc> of <nc>information concept objects</nc> and <nc>one or more actual links</nc> between <nc>the information concept objects</nc>; receiving, by <nc>the data processing system</nc>, <nc>an indication</nc> of <nc>at least a selected information concept object</nc> for <nc>which</nc> <nc>a hypothetical ontological link</nc> is to be evaluated, wherein <nc>the hypothetical ontological link</nc> is <nc>a potential link</nc> <nc>that</nc> is not already present as <nc>an actual link</nc> in <nc>the ontology</nc>; automatically generating, by <nc>the data processing system</nc>, <nc>one or more natural language questions</nc> for <nc>processing</nc> by <nc>a Question Answering (QA) system pipeline</nc> based on <nc>at least an identification</nc> of <nc>a type</nc> of <nc>the selected information concept object</nc>; <nc>processing</nc>, by <nc>the QA system pipeline</nc>, <nc>the one or more natural language questions</nc> to generate <nc>answer results</nc>; calculating, by <nc>the data processing system</nc>, <nc>a score</nc> for <nc>the hypothetical ontological link</nc> based on <nc>the answer results</nc>; and outputting, by <nc>the data processing system</nc>, <nc>information</nc> associated with <nc>the hypothetical ontological link</nc> based on <nc>the score</nc> for <nc>the hypothetical ontological link</nc>, wherein calculating <nc>a score</nc> for <nc>the hypothetical ontological link</nc> comprises: calculating <nc>a score</nc> for <nc>each answer result</nc> in <nc>the generated answer results</nc>; generating <nc>a weighted score</nc> for <nc>each</nc> of <nc>the answer results</nc>; and combining <nc>the weighted scores</nc> for <nc>each</nc> of <nc>the answer results</nc> to generate <nc>the score</nc> for <nc>the hypothetical ontological link</nc>.
| 3
|
3. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: automatically identifying, by <nc>the data processing system</nc>, <nc>the hypothetical ontological link</nc> to be evaluated based on <nc>the indication</nc> of <nc>at least a selected information concept object</nc>.
|
5488719
|
07814552
|
1. An article of <nc>manufacture</nc> for <nc>use</nc> in <nc>a character recognition system</nc> <nc>that</nc> includes <nc>a processor</nc>; the article comprising: <nc>a data storage medium</nc>; and string <nc>data</nc> stored by <nc>the data storage medium</nc>; the string data comprising <nc>two or more data units</nc>, <nc>each</nc> of <nc>which</nc> can be accessed by <nc>a processor</nc> of <nc>a character recognition system</nc>; <nc>the data units</nc> including, for <nc>each</nc> of <nc>a set</nc> of <nc>two or more acceptable strings</nc> of <nc>characters</nc>, <nc>a respective sequence</nc> of <nc>data units</nc> <nc>that</nc> <nc>the processor</nc> can access using <nc>character data</nc> indicating <nc>character types</nc> of <nc>the string's characters</nc>; <nc>the set</nc> of <nc>acceptable strings</nc> including <nc>a first string</nc> <nc>that</nc> is in <nc>a first subset</nc> of <nc>categories</nc>, <nc>each category</nc> in the first subset being one of <nc>a set</nc> of <nc>two or more categories</nc>; <nc>the first string's sequence</nc> of <nc>data units</nc> including <nc>a respective ending subsequence</nc> of <nc>data units</nc> <nc>that</nc> <nc>the processor</nc> can access at <nc>the end</nc> of <nc>the first string's sequence</nc> and <nc>use</nc> to obtain <nc>first string ending data</nc> indicating that <nc>the first string</nc> is one of <nc>the acceptable strings</nc> and indicating <nc>the first subset</nc> of <nc>categories</nc>; <nc>the first string's ending subsequence</nc> including: <nc>acceptance information</nc> indicating that <nc>a string</nc> at <nc>the end</nc> of <nc>whose respective sequence</nc> <nc>the processor</nc> can access <nc>the ending subsequence</nc> is one of <nc>the set</nc> of <nc>acceptable strings</nc>; and <nc>category</nc> set <nc>information</nc> indicating <nc>the first subset</nc> of <nc>categories</nc>; <nc>the first subset</nc> of <nc>categories</nc> including at least one of <nc>the set</nc> of <nc>categories</nc>.
| 7
|
7. <nc>The article</nc> of <nc>claim</nc> 1 in <nc>which</nc> <nc>the acceptance information</nc> includes <nc>an acceptance data unit</nc> in <nc>the first string's ending subsequence</nc>; <nc>the acceptance data unit</nc> having <nc>a value</nc> indicating that <nc>a string</nc> at <nc>the end</nc> of <nc>whose respective sequence</nc> <nc>the processor</nc> can access <nc>the ending subsequence</nc> is one of <nc>the acceptable strings</nc>.
|
8700673
|
13398792
|
1. <nc>A method</nc> comprising: accessing <nc>metadata items</nc> <nc>that</nc> specify <nc>at least structural constraints</nc> on <nc>data objects</nc> within <nc>a data repository</nc>, <nc>the metadata items</nc> being separate from <nc>the data objects</nc> for <nc>which</nc> <nc>the metadata items</nc> specify <nc>the structural constraints</nc>; generating <nc>an index</nc>, <nc>the index</nc> mapping <nc>the metadata items</nc> to <nc>terms</nc> associated with <nc>the metadata items</nc>; generating <nc>a graph</nc> describing <nc>relationships</nc> between <nc>each</nc> of <nc>the metadata items</nc>; receiving <nc>a search request</nc> comprising <nc>at least one or more search terms</nc>; based on <nc>the one or more search terms</nc> and <nc>the index</nc>, locating <nc>a candidate set</nc> of <nc>the metadata items</nc>; performing <nc>a link analysis</nc> of <nc>the graph</nc> to determine <nc>a relationship score</nc> for <nc>each particular metadata item</nc> in <nc>at least the candidate</nc> set of <nc>metadata items</nc>; for <nc>each particular metadata item</nc> in <nc>the candidate set</nc> of <nc>the metadata items</nc>, calculating <nc>a ranking score</nc> based at least on <nc>the relationship score</nc> for <nc>the particular metadata item</nc>; generating a ranked result set based on comparing <nc>the ranking scores</nc> for <nc>the candidate set</nc> of <nc>metadata items</nc>, the ranked result set including <nc>at least one metadata item</nc> in <nc>the candidate set</nc>; providing <nc>information</nc> indicating <nc>the ranked result</nc> set in <nc>response</nc> to <nc>the search request</nc>; wherein <nc>the method</nc> is performed by <nc>one or more computing devices</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the graph</nc> comprises <nc>nodes</nc> and <nc>edges</nc>, <nc>the nodes</nc> corresponding to <nc>the metadata items</nc> and <nc>the edges</nc> corresponding to <nc>the relationships</nc>, wherein <nc>at least some</nc> of <nc>the edges</nc> have <nc>different weights</nc> corresponding to <nc>different types</nc> of <nc>the relationships</nc> <nc>that</nc> are represented by <nc>the edges</nc>.
|
5404510
|
07886751
|
1. For <nc>a database</nc> having <nc>plural tables</nc>, <nc>each</nc> of <nc>plural columns</nc>, stored in <nc>memory</nc>, <nc>a processor controlled method</nc> of generating <nc>indexes</nc> comprising: identifying <nc>importance values</nc> for <nc>individual database requests</nc>; and designing <nc>indexes</nc> to <nc>tables</nc> for <nc>the requests</nc> by identifying, by <nc>order</nc> of <nc>request importance</nc>, <nc>candidate indexes</nc> for <nc>requests</nc>, searching <nc>previously identified indexes</nc> for <nc>an index</nc> <nc>that</nc> is similar to <nc>each candidate index</nc>, and building upon <nc>previously identified indexes</nc> by reusing <nc>existing indexes</nc> and modifying <nc>existing indexes</nc> based upon <nc>match</nc> between <nc>indexes</nc> and upon <nc>importance values</nc> of <nc>requests</nc> for <nc>which</nc> <nc>the indexes</nc> are created.
| 18
|
18. <nc>A method</nc> as claimed in <nc>claim</nc> 1 further comprising <nc>the step</nc> of identifying <nc>index retrieval modes</nc> based on <nc>context operators</nc>.
|
7640162
|
11011811
|
1. <nc>A system</nc> for synchronizing <nc>a natural language input element</nc> and <nc>a graphical user interface element</nc>, <nc>the system</nc> comprising: <nc>the natural language input element</nc>, <nc>wherein the natural language input element displays</nc>, via <nc>a display device</nc>, <nc>an indication</nc> of <nc>a user natural language input</nc> on <nc>a user interface</nc>, wherein <nc>the displayed indication</nc> of <nc>the user natural language input</nc> is <nc>a representation</nc> of <nc>words</nc> included within <nc>an initial natural language input</nc> received from <nc>a user</nc>, and <nc>wherein the displayed indication</nc> of <nc>the user natural language input</nc> is displayed in <nc>an editable format</nc> <nc>that</nc> allows <nc>the user</nc> to modify <nc>the representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc> by adding <nc>words</nc> to or removing <nc>words</nc> from <nc>the representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc>; <nc>the graphical user interface element</nc>, <nc>wherein the graphical user interface element displays</nc>, via <nc>a display device</nc>, on <nc>the user interface</nc> simultaneously with <nc>the display</nc> of <nc>the indication</nc> of <nc>the user natural language input</nc>, <nc>an indication</nc> of <nc>a user graphical interface input</nc>, wherein <nc>the displayed indication</nc> of <nc>the user graphical interface input</nc> is <nc>an automatically generated control box representation</nc> of <nc>the initial natural language input</nc> received from <nc>the user</nc>, <nc>the displayed indication</nc> of <nc>the user graphical interface input</nc> and <nc>the displayed indication</nc> of <nc>the user natural language input</nc> being <nc>two different representations</nc> of <nc>the same initial natural language input</nc> received from <nc>the user</nc>, and <nc>wherein the displayed indication</nc> of <nc>the user graphical interface input</nc> is <nc>an editable format</nc> <nc>that</nc> allows <nc>the user</nc> to modify <nc>the automatically generated control box representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc>; <nc>a synchronization engine</nc> <nc>that</nc> monitors, via <nc>a processor</nc>, <nc>user interactions</nc> with <nc>the natural language input element</nc> and <nc>the graphical user interface element</nc> and automatically synchronizes <nc>the natural language input element</nc> and <nc>the graphical user interface element</nc> responsive to <nc>the user interactions</nc>, wherein <nc>the synchronization engine</nc> automatically synchronizes <nc>the natural language input element</nc> and <nc>the graphical user interface element</nc> responsive to <nc>the user interactions</nc> by automatically changing <nc>the displayed indication</nc> of <nc>the user graphical interface input</nc> so as to automatically modify <nc>the automatically generated control box representation</nc> so as to alter <nc>the automatically generated control box representation</nc> to incorporate <nc>revised logic</nc> <nc>that</nc> is consistent with <nc>an edit</nc> made to <nc>the representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc>, <nc>the edit</nc> being <nc>a word</nc> added to or removed from <nc>the representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc>; and <nc>a restatement engine</nc> <nc>that</nc> monitors <nc>user interactions</nc> with <nc>the graphical user interface element</nc> and utilizes <nc>a computer processor</nc> <nc>that</nc> is <nc>a component</nc> of <nc>a computing device</nc> to automatically compose <nc>a natural language input representative</nc> of <nc>a modification</nc> to <nc>the automatically generated control box representation</nc> of <nc>the initial natural language input</nc> received from <nc>the user</nc>, wherein <nc>the modification</nc> alters <nc>the logic</nc> of <nc>the displayed representation</nc>, and wherein <nc>the synchronization engine</nc> updates <nc>the natural language element</nc> so as to automatically respond to <nc>the modification</nc> to <nc>the automatically generated representation</nc> by automatically substituting <nc>a display</nc> of <nc>the automatically composed natural language input</nc> for <nc>the displayed representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc> received from <nc>the user</nc>, wherein <nc>the automatically composed natural language input</nc> is <nc>an automatically generated re</nc><nc>-</nc><nc>statement</nc> of <nc>the displayed representation</nc> of <nc>words</nc> included within <nc>the initial natural language input</nc> received from <nc>the user</nc> with <nc>modifications</nc> being automatically made to <nc>the displayed representation</nc> of <nc>words</nc> based directly on <nc>the modification</nc> to <nc>the displayed representation</nc>.
| 9
|
9. <nc>The system</nc> of <nc>claim</nc> 1 , wherein <nc>the displayed indication</nc> of <nc>the user graphical interface input</nc> is <nc>an automatically generated control box representation</nc> <nc>that</nc> includes <nc>a series</nc> of <nc>text</nc> and check <nc>boxes</nc> <nc>that</nc> <nc>each</nc> contain <nc>an entry</nc> such that, collectively, <nc>the text and check boxes</nc> represent <nc>multiple elements</nc> of <nc>the initial natural language input</nc> received from <nc>the user</nc>.
|
10062300
|
14983542
|
1. <nc>A modular learning device</nc> comprising of: <nc>a base board</nc> having <nc>a frame</nc> with <nc>plurality</nc> of <nc>sides</nc>, <nc>a complementary joining construction</nc> on <nc>sides</nc> of <nc>the base board</nc>, <nc>a front plane</nc>, <nc>a rear plane</nc>, <nc>a network</nc> of <nc>tracks</nc> having <nc>a spinal track</nc>, <nc>several dormant tracks</nc>, <nc>several solution tracks</nc>, <nc>several parking tracks</nc>, <nc>the dormant tracks</nc> and <nc>the parking tracks</nc> and <nc>the solution tracks</nc> are connected to <nc>the spinal track</nc>, <nc>a recess</nc>, wherein <nc>some parking tracks</nc> are also <nc>solution tracks</nc>; <nc>a plurality</nc> of <nc>sliding blocks</nc> having <nc>a tread</nc>, <nc>a stopper</nc> and <nc>a shaft</nc>, <nc>the shaft</nc> having <nc>a first end</nc> and a second end; and <nc>a plurality</nc> of <nc>wisdom cards</nc> having <nc>an identifier section</nc>, <nc>a margin</nc> having up to <nc>same number</nc> of <nc>sides</nc> as <nc>the base board</nc>, <nc>a plurality</nc> of <nc>zones</nc> in <nc>the margin</nc>, <nc>a middle area</nc> and <nc>corresponding solutions</nc> in <nc>the zones</nc> randomly, <nc>a solution code</nc> for <nc>problems</nc> or <nc>situations</nc>, <nc>each wisdom card</nc> containing <nc>different problems</nc> or <nc>situations</nc> on <nc>a plurality</nc> of <nc>subjects</nc>, <nc>an indicator</nc> for <nc>each problem</nc> or <nc>situation</nc>, wherein <nc>the indicator</nc> is <nc>a hint</nc> or <nc>a prompt</nc> to arrive at <nc>correct provided solution</nc>; <nc>different base boards</nc> unifiable by <nc>the complementary joining construction</nc> on <nc>the sides</nc> of <nc>the base board</nc>, <nc>the plurality</nc> of <nc>wisdom cards</nc> mounted, either one at <nc>a time</nc>, or in <nc>combination</nc>, one in <nc>each recess</nc> of <nc>the base board</nc>, <nc>the plurality</nc> of <nc>sliding blocks</nc> mounted on <nc>the base board</nc> with <nc>the head</nc> on <nc>the front plane</nc> of <nc>the base board</nc> and <nc>the stopper</nc> on <nc>the rear plane</nc> of <nc>the base board</nc>, <nc>the plurality</nc> of <nc>sliding blocks</nc> can move freely in <nc>the network</nc> of <nc>tracks</nc> and cannot get dislodged unless <nc>the stopper</nc> is separated or manoeuvred using <nc>a minimum force</nc>.
| 9
|
9. <nc>The modular learning device</nc> as claimed in <nc>claim</nc> 1 , wherein <nc>the head</nc> of <nc>the sliding block</nc> has <nc>a top surface</nc> having <nc>a distinct identification</nc>.
|
8892423
|
13004019
|
1. <nc>A computer-implemented method</nc> for generating <nc>examples</nc> for <nc>electronic dictionaries</nc> to serve as <nc>an aid</nc> to <nc>translation</nc> between <nc>languages</nc> performed by <nc>one or more processors</nc>, <nc>the method</nc> comprising: creating <nc>an electronic dictionary example</nc> by: acquiring <nc>at least one dictionary entry</nc> comprising <nc>a headword</nc> <nc>W j</nc> in <nc>a source language</nc> and <nc>at least one translation T j1</nc> , <nc>T j2</nc> , . . . <nc>T jn</nc> for <nc>the headword</nc> <nc>W j</nc> in <nc>a target language</nc>; generating <nc>a first set</nc> comprising <nc>possible forms</nc> for <nc>the headword</nc> <nc>W j</nc> in <nc>the source language</nc> and <nc>a second set</nc> comprising <nc>possible forms</nc> for <nc>each translation</nc> <nc>T j1</nc> , <nc>T j2</nc> , . . . <nc>T jn</nc> in <nc>the target language</nc>; searching <nc>a corpus</nc> of <nc>translations</nc>, where <nc>the corpus</nc> of <nc>translations</nc> is <nc>a preexisting corpus</nc> of <nc>translation sentence pairs</nc><nc>, each translation sentence pair</nc> comprising <nc>a first sentence</nc> in <nc>the source language</nc> and <nc>a second sentence</nc> in <nc>the target language</nc>, where <nc>the first sentence</nc> is <nc>a translation</nc> of <nc>the second sentence</nc>, and <nc>the searching</nc> includes searching <nc>at least one first sentence</nc> in <nc>the source language</nc> included in <nc>the corpus</nc> of <nc>translations</nc> and searching <nc>at least one second sentence</nc> in <nc>the target language</nc> in <nc>the corpus</nc> of <nc>translations</nc>; identifying in <nc>the corpus</nc> of <nc>translations</nc> <nc>at least one translation sentence pair</nc>, from <nc>either the searching</nc> of <nc>the at least one first sentence</nc> in <nc>the source language</nc> or <nc>the searching</nc> of <nc>the at least one second sentence</nc> in <nc>the target language</nc>, <nc>that</nc> consists of <nc>the first sentence</nc> <nc>that</nc> incorporates <nc>the headword</nc> <nc>W j</nc> , or one of <nc>its generated forms</nc>, and the second sentence <nc>that</nc> incorporates <nc>the translation T jn</nc> or one of <nc>its generated forms</nc>; and providing <nc>the at least one translation sentence pair</nc> to <nc>a user</nc>.
| 14
|
14. <nc>The computer-implemented method</nc> of <nc>claim</nc> 1 , wherein <nc>the at least one translation sentence pair</nc> comprises <nc>the headword</nc> <nc>W j</nc> in <nc>a first part</nc> and one of <nc>the translation T jn</nc> , <nc>its generated form</nc>, and <nc>a semantically expanded form</nc>, in <nc>a second part</nc>.
|
8326850
|
12837801
|
1. <nc>A data converting apparatus</nc> comprising: <nc>a storage unit</nc> <nc>that</nc> <nc>stores</nc> encoded <nc>meta-definition information</nc> <nc>that</nc> assigns <nc>a metadata code</nc> as <nc>a unique code</nc> to <nc>an element</nc> making up <nc>metadata</nc> in <nc>meta-definition information</nc> <nc>that</nc> defines <nc>metadata</nc> indicative of <nc>a property</nc> related to <nc>data</nc> of <nc>a conversion source</nc> and <nc>a conversion destination</nc>, a data converting <nc>function</nc> <nc>that</nc> converts <nc>conversion source data</nc> having <nc>a property</nc> prescribed by <nc>the metadata</nc> for <nc>the conversion source</nc> into <nc>conversion destination data</nc> having <nc>a property</nc> prescribed by <nc>the metadata</nc> for <nc>the conversion destination</nc>, <nc>a conversion rule table</nc> <nc>that</nc> assigns the data converting <nc>function</nc> according to <nc>a combination</nc> of <nc>a metadata code</nc> for <nc>the conversion source</nc> and <nc>a metadata code</nc> for <nc>the conversion destination</nc>, and <nc>a conversion rule</nc> <nc>that</nc> correlates with <nc>each</nc> of <nc>the conversion rule tables</nc>, <nc>a relevant metadata code</nc> as <nc>a conversion rule code</nc>; <nc>an input unit</nc> <nc>that</nc> receives <nc>input</nc> of <nc>data</nc> to be converted; <nc>a detecting unit</nc> <nc>that</nc> refers to <nc>the encoded meta-definition information</nc> stored in <nc>the storage unit</nc> and detects <nc>the metadata codes</nc> for <nc>the conversion source</nc> and <nc>the conversion destination</nc> for <nc>which</nc> the conversion rule code matches between <nc>the conversion source</nc> and <nc>the conversion destination</nc>; <nc>a determining unit</nc> <nc>that</nc> determines whether <nc>the detected metadata codes</nc> for <nc>the conversion source</nc> and for <nc>the conversion destination match</nc>; a converting function specifying <nc>unit</nc> <nc>that</nc>, by referring to <nc>a conversion rule</nc> stored in <nc>the storage unit</nc> and based on <nc>the determination result</nc> obtained by <nc>the determining unit</nc>, specifies <nc>the data</nc> converting <nc>function</nc>, according to <nc>the combination</nc> of <nc>the metadata code</nc> for <nc>the conversion source</nc> and <nc>the metadata code</nc> for <nc>the conversion destination</nc>; and <nc>a converting unit</nc> <nc>that</nc>, by using <nc>the data converting function</nc> specified by <nc>the converting function specifying unit</nc>, converts <nc>the conversion source data</nc>, <nc>which</nc> is <nc>the data</nc> to be converted, to have <nc>a property</nc> prescribed by <nc>metadata</nc> for <nc>the conversion destination</nc>.
| 2
|
2. <nc>The data</nc> converting <nc>apparatus</nc> according to <nc>claim</nc> 1 , wherein <nc>when matching</nc> of <nc>the metadata codes</nc> is determined by <nc>the determining unit</nc>, <nc>the converting function specifying unit</nc> does not specify the data converting <nc>function</nc> and <nc>the converting unit</nc> does not convert <nc>the data</nc> to be converted.
|
8706491
|
12862001
|
1. <nc>A method</nc> of training <nc>an information extraction system</nc> to extract <nc>information</nc> from <nc>a natural language input</nc>, comprising: initializing <nc>a structured language model</nc> with <nc>syntactically annotated training data</nc>, <nc>the annotated training data</nc> including <nc>a parse tree</nc> for <nc>a sentence</nc> having <nc>syntactic labels</nc> comprising <nc>a frame label</nc> indicating <nc>an overall action</nc> being referred to by <nc>the sentence and slot labels</nc> identifying <nc>attributes</nc> of <nc>the action</nc>; training <nc>the structured language model</nc> by generating <nc>parses</nc> with <nc>the initialized structured language model</nc> using <nc>annotated training data</nc> <nc>that</nc> has <nc>semantic constituent labels</nc> with <nc>semantic constituent boundaries</nc> identified, wherein <nc>the structured language model</nc> is trained as <nc>a match constrained parser</nc> <nc>which</nc> generates <nc>a set</nc> of syntactic parses for <nc>a given word string</nc> that <nc>all</nc> match <nc>the constituent boundaries</nc> specified by <nc>the semantic parse</nc>, by determining whether <nc>unlabeled constituents</nc> <nc>that</nc> define <nc>the semantic parse</nc> are included in <nc>a set</nc> of <nc>constituents</nc> <nc>that</nc> define <nc>the syntactic parse</nc>, <nc>wherein any parses</nc> <nc>that</nc> do not match <nc>the constituent boundaries</nc> are discarded; replacing <nc>the syntactic labels</nc> in <nc>the parse tree</nc> with <nc>joint syntactic and semantic labels</nc> based on <nc>the generated parses</nc> excluding <nc>the discarded parses</nc>; and retraining <nc>the structured language model</nc> in <nc>which</nc> <nc>the structured language model</nc> generates <nc>parses</nc> <nc>that</nc> are constrained to identically match <nc>the semantic constituent labels</nc> of <nc>the joint syntactic and semantic labels</nc> and constrained to match <nc>all</nc> of <nc>the semantic constituent boundaries</nc>.
| 5
|
5. <nc>The method</nc> of <nc>claim</nc> 1 <nc>wherein generating parses</nc> comprises: generating syntactic parses with <nc>syntactic labels</nc>, wherein the syntactic parses are constrained to match <nc>the semantic constituent boundaries</nc>; and generating <nc>semantic parses</nc> with <nc>semantic labels</nc>, wherein <nc>the semantic parses</nc> are constrained to match <nc>the semantic constituent labels</nc> in <nc>the annotated training data</nc>.
|
5504902
|
08160957
|
1. <nc>A method</nc> of generating <nc>a control program</nc> utilizing <nc>a computer system</nc>, <nc>the control program</nc> executable by <nc>a controller</nc> to control <nc>the operation</nc> of <nc>a one or more devices</nc>, <nc>the method</nc> comprising <nc>the steps</nc> of: receiving <nc>one or more instructions</nc> of <nc>a source program</nc> in <nc>a high-level text-based language</nc>; converting at least one of <nc>the one or more instructions</nc> into <nc>first corresponding instructions</nc> in <nc>a ladder-based language</nc>; editing <nc>the first corresponding instructions</nc> of <nc>the source program</nc> in <nc>the ladder-based language</nc> to form <nc>an edited source program</nc>; and compiling <nc>the edited source program</nc> to form <nc>the control program</nc>.
| 9
|
9. <nc>The method</nc> of <nc>claim</nc> 1 wherein <nc>the computer system</nc> is operated by <nc>a user</nc> and wherein <nc>the method</nc> further comprises <nc>the steps</nc> of: receiving <nc>a second signal</nc> generated by <nc>the user</nc>; and converting <nc>the first corresponding instructions</nc> in <nc>the ladder-based language</nc>, as edited, to <nc>third corresponding instructions</nc> in <nc>the high-level text-based language</nc> only if <nc>the second signal</nc> is received.
|
8131536
|
11998663
|
1. <nc>A method</nc> for automatically translating <nc>a document</nc> from <nc>a first language</nc> to <nc>a second language</nc> comprising: receiving <nc>the document</nc> in <nc>the first language</nc>; processing <nc>the document</nc> to extract <nc>elements</nc> of <nc>information</nc>; determining, using <nc>a processor</nc>, <nc>a plurality</nc> of <nc>potential translations</nc> for <nc>each</nc> of <nc>the extracted elements</nc> of <nc>information</nc> using <nc>a first translation process</nc> and <nc>a likelihood value</nc> for <nc>each</nc> of <nc>the potential translations</nc> of <nc>the elements</nc> of <nc>information</nc>; determining <nc>a plurality</nc> of <nc>potential translations</nc> of <nc>a remainder</nc> of <nc>the document</nc> using <nc>a second, different translation process</nc> and <nc>a likelihood value</nc> for <nc>each potential remainder translation</nc>; generating <nc>a plurality</nc> of <nc>combinations</nc> by combining <nc>a plurality</nc> of <nc>the potential translations</nc> of <nc>the elements</nc> of <nc>information</nc> with <nc>a plurality</nc> of <nc>the potential remainder translations</nc>; determining <nc>a likelihood value</nc> for <nc>respective ones</nc> of <nc>a plurality</nc> of <nc>the combinations</nc> based on <nc>a model</nc> of <nc>the second language</nc> and <nc>corresponding likelihood values</nc> of <nc>each</nc> of <nc>the potential element</nc> of <nc>information translations</nc> and <nc>remainder translations</nc> included in <nc>the respective combinations</nc>; and forming <nc>a translated version</nc> of <nc>the document</nc> based on <nc>the likelihood values</nc> of <nc>the combinations</nc>.
| 8
|
8. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the second translation process</nc> includes employing <nc>a statistical translation process</nc>.
|
8112257
|
11831629
|
1. <nc>A method</nc> comprising: receiving <nc>a definition</nc> of <nc>an association link</nc> between <nc>a data object</nc> of <nc>a business process model</nc> and <nc>a modeling element</nc> of <nc>the business process model</nc>, <nc>the modeling element</nc> representing <nc>an activity</nc>; presenting <nc>a plurality</nc> of <nc>domain elements</nc> to <nc>a user</nc>, wherein presenting <nc>the plurality</nc> of <nc>domain elements</nc> include presenting <nc>a first plurality</nc> of <nc>domain elements</nc> and <nc>a second plurality</nc> of <nc>domain elements</nc>, <nc>each</nc> of the first and <nc>the second plurality</nc> of <nc>domain elements</nc> being included in <nc>a domain ontology</nc>; receiving a first and <nc>a second identification</nc> of <nc>a domain element</nc> from the first and <nc>the second plurality</nc> of <nc>domain elements</nc>, the first and <nc>the second identification</nc> of <nc>the domain element</nc> being <nc>a user selection</nc> of <nc>the domain element</nc> from the first and <nc>the second plurality</nc> of <nc>domain elements</nc> wherein the first and <nc>the second plurality</nc> of <nc>domain elements</nc> are identified using <nc>context information</nc> selected from <nc>user input</nc> and <nc>adjacent business modeling components</nc>; and semantically annotating, using <nc>a processor</nc>, <nc>the definition</nc> of <nc>the association link</nc> between <nc>the data object</nc> and <nc>the modeling element</nc> using <nc>the domain element</nc>.
| 3
|
3. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising receiving <nc>a definition</nc> of <nc>a business process modeling element</nc> of <nc>the business process model</nc>.
|
8490050
|
12104439
|
1. <nc>A system</nc> configured to generate <nc>a user interface</nc> comprising: <nc>a processor</nc> configured to execute <nc>the following computer executable components</nc>; <nc>a schema generator</nc> configured to receive <nc>metadata</nc> for <nc>a reference object</nc> and further configured to generate from <nc>the metadata</nc> <nc>a service-neutral interface schema</nc> for <nc>reference object access points</nc> determined to have <nc>communicative intent</nc>, wherein <nc>the service-neutral interface schema</nc> is <nc>an abstraction</nc> of <nc>the reference object access points</nc> identified from <nc>an analysis</nc> of <nc>the metadata</nc>, wherein <nc>the schema generator</nc> is configured to determine if <nc>an access point</nc> to <nc>the reference object</nc> implies <nc>communicative intent</nc> by determining whether <nc>the access point</nc> has <nc>both an input</nc> and <nc>an output</nc> <nc>that</nc> complies with <nc>a provided service-specific schema pattern</nc>, wherein <nc>the output</nc> is produced as <nc>a result</nc> of <nc>the input</nc>; and <nc>a user interface generator</nc> configured to receive <nc>the service-neutral interface schema</nc> and <nc>a transform</nc> directed to <nc>a particular target platform</nc> and further configured to generate <nc>a service-specific user interface</nc> from <nc>the transform</nc> and <nc>the service neutral interface schema</nc> generated from <nc>the metadata</nc>.
| 5
|
5. <nc>The system</nc> of <nc>claim</nc> 1 , wherein <nc>the user interface generator</nc> is configured to generate <nc>a user interface</nc> targeted to <nc>a Windows Presentation Foundation (WPF) platform</nc> from <nc>a web services communication class</nc>.
|
9934289
|
14995724
|
1. <nc>A method</nc> for performing <nc>fuzzy full text search</nc>, comprising: receiving <nc>a plurality</nc> of <nc>search terms</nc>, <nc>the plurality</nc> of <nc>search terms</nc> comprising <nc>a first search term</nc> and <nc>a second search term</nc>; identifying <nc>paths</nc> of <nc>an inverted token</nc> <nc>next valid character tree</nc> including <nc>a plurality</nc> of <nc>nodes</nc> including <nc>a root node</nc> and <nc>at least one leaf node</nc> for <nc>at least one document</nc>, <nc>wherein one or more paths</nc> from <nc>the root node</nc> to <nc>the at least one leaf node</nc> corresponds to at least one of <nc>(a</nc>) at least one first token for <nc>the first search term</nc> or <nc>(b) at least one second token</nc> for <nc>the second search term</nc>; computing, using <nc>a processor</nc>, <nc>a plurality</nc> of <nc>document error lists</nc> corresponding to <nc>the plurality</nc> of <nc>search terms</nc> matched with <nc>the one or more paths</nc> <nc>that</nc> reach <nc>the at least one leaf node</nc> in <nc>the next valid character tree</nc>, <nc>the plurality</nc> of <nc>document error lists</nc> comprising <nc>a first document error list</nc> and <nc>a second document error list</nc>, wherein (a) <nc>the first document error list</nc> comprises <nc>one or more instances</nc> of <nc>first error list information</nc>, <nc>each instance</nc> of <nc>first error list information</nc> comprising <nc>a first document identifier</nc> and <nc>a corresponding first error distance</nc>, (b) <nc>the second error list</nc> comprises <nc>one or more instances</nc> of <nc>second error list information</nc>, <nc>each instance</nc> of <nc>second error list information</nc> comprising <nc>a second document identifier</nc> and <nc>a corresponding second error distance</nc>, (c) <nc>the first document identifier</nc> is associated with <nc>a document</nc> <nc>that</nc> comprises <nc>at least one instance</nc> of the at least one first token, and (d) <nc>the second document identifier</nc> is associated with <nc>a document</nc> <nc>that</nc> comprises <nc>at least one instance</nc> of <nc>the at least one second token</nc>; comparing <nc>the plurality</nc> of <nc>document error lists</nc> to identify <nc>an instance</nc> of <nc>first error list information</nc> and <nc>an instance</nc> of <nc>second error list information</nc>, wherein <nc>(a) the first document identifier</nc> and <nc>the second document identifier</nc> are <nc>a common document identifier</nc> and (b) at least one of (<nc>i</nc>) <nc>the first error distance</nc> satisfies <nc>an error distance threshold</nc> and <nc>the second error distance</nc> satisfies <nc>the error distance threshold</nc> or <nc>(ii</nc>) <nc>the sum</nc> of <nc>the first error distance</nc> and <nc>the second error distance</nc> satisfies <nc>the error distance threshold</nc>; adding <nc>the first document identifier</nc> to <nc>a result set</nc>; and providing <nc>the result</nc> set of <nc>document identifiers</nc>.
| 7
|
7. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising: storing <nc>an index</nc> including <nc>the next valid character tree</nc>.
|
7533034
|
10044779
|
1. <nc>A computer</nc> implemented <nc>method</nc> for providing through <nc>a computer network</nc> to <nc>business management</nc> <nc>a plan</nc> for implementing <nc>a user's suggestion</nc> for <nc>business improvement</nc>, the method comprising: in <nc>a first computer process</nc>, causing <nc>presentation</nc> to <nc>a user</nc> seeking to submit <nc>a suggestion</nc> for <nc>business improvement</nc>, <nc>a series</nc> of <nc>two or more templates</nc> for entering <nc>a structured response</nc> on <nc>a terminal device</nc>, wherein one of <nc>the templates</nc> presented to <nc>the user</nc> allows <nc>the user</nc> to characterize <nc>the type</nc> of <nc>suggestion</nc> as falling into at least one of <nc>a plurality</nc> of <nc>categories</nc> selected from <nc>a group</nc> of <nc>cost saving</nc>, <nc>revenue generation</nc>, <nc>quality improvement</nc>, <nc>safety improvement</nc>, <nc>customer service improvement</nc>, <nc>development</nc> of <nc>a new product</nc>, <nc>policy change</nc> and <nc>advertising</nc> or <nc>corporate slogan</nc>; receiving over <nc>a computer network</nc> <nc>the structured response</nc>, entered into <nc>the two or more templates</nc> from <nc>the user</nc>, wherein <nc>the structured response</nc> includes <nc>a characterization</nc> of <nc>the type</nc> of <nc>suggestion</nc> entered into <nc>one or more templates</nc> by <nc>the user</nc> and <nc>a server</nc> logically selects at least one of <nc>the templates</nc> presented to <nc>the user</nc> according to <nc>the type</nc> of <nc>suggestion</nc> characterized by <nc>the user</nc>; and in <nc>a second computer process</nc>, determining <nc>the network routing</nc> of <nc>data</nc> from <nc>the structured response</nc> to <nc>business management</nc> based upon <nc>entries</nc> of <nc>the response</nc> in <nc>one or more templates</nc>.
| 14
|
14. <nc>The method</nc> according <nc>to claim</nc> 1 wherein determining <nc>the network routing</nc> is dependent in <nc>part</nc> on whether <nc>the user</nc> indicates that <nc>the suggestion</nc> is <nc>a team suggestion</nc>.
|
9697830
|
14750604
|
1. <nc>A computer program product</nc> comprising <nc>a non-transitory computer readable storage medium</nc> having <nc>program instructions</nc> embodied therewith, <nc>the program instructions</nc> executable by <nc>a processor</nc> to cause <nc>the processor</nc> to perform <nc>a method</nc> comprising: generating <nc>a time-marked word list</nc> of <nc>an automatic speech recognition system</nc>, wherein generating <nc>the time-marked word list</nc> comprises converting <nc>an indexing structure</nc> into <nc>an output</nc>, and wherein <nc>the time-marked word list</nc> comprises <nc>the output</nc>; generating <nc>an index</nc> from <nc>the time-marked word list</nc>, wherein generating <nc>the index</nc> comprises creating <nc>a word loop weighted finite state transducer</nc> for <nc>each utterance</nc>, <nc>i</nc>, of <nc>a plurality</nc> of <nc>utterances</nc>; receiving <nc>a plurality</nc> of <nc>keyword queries</nc>; and searching <nc>the index</nc> for <nc>a plurality</nc> of <nc>keyword hits</nc>; wherein <nc>the word</nc> <nc>loop</nc> weighted finite state transducer for <nc>each utterance</nc>, <nc>i</nc>, of <nc>the plurality</nc> of <nc>utterances</nc>, includes <nc>S i</nc> as <nc>a start node</nc>, <nc>E i</nc> as <nc>an end node</nc>, without <nc>a start node</nc> or <nc>an end node</nc> between <nc>S i</nc> and <nc>E i</nc> and <nc>a plurality</nc> of <nc>arcs</nc> connected between <nc>an S i</nc> to <nc>E</nc> <nc>i</nc> pair for <nc>each utterance</nc>, <nc>the plurality</nc> of <nc>arcs</nc> corresponding to <nc>each word label</nc>, start and end time, and posterior probability in <nc>the time-marked word list</nc>.
| 3
|
3. <nc>The computer program product</nc> according to <nc>claim</nc> 1 , wherein generating <nc>the index</nc> further comprises: creating <nc>a new start node</nc>, <nc>S</nc>, <nc>that</nc> is connected to <nc>each S i</nc> by <nc>zero-cost arcs</nc> with <nc>input label epsilon</nc> and <nc>output label</nc> <nc>i</nc>; and creating <nc>a new end node</nc>, <nc>E</nc>, <nc>that</nc> is connected to <nc>each E i</nc> by <nc>zero-cost epsilon-arcs</nc>, wherein <nc>each S i</nc> to <nc>E</nc> <nc>i</nc> pair is connected by <nc>the plurality</nc> of <nc>arcs</nc>.
|
9853818
|
15132769
|
1. <nc>A system</nc> for authenticating <nc>an electronic document</nc> having <nc>a digital signing signature</nc> enabling <nc>a relying party</nc> <nc>that</nc> receives <nc>the digitally signed electronic document</nc> to evaluate <nc>a risk</nc> of relying on <nc>the digitally signed electronic document</nc>, comprising: <nc>an electronic network</nc> configured for <nc>electronic communication</nc> among <nc>a plurality</nc> of <nc>communication devices</nc> <nc>each</nc> associated with <nc>a respective computer system</nc> operated by <nc>a certification authority</nc>, by <nc>a signature authority</nc>, and by <nc>a signing party</nc>; <nc>the certification authority computer system</nc> configured for generating by <nc>the certification authority</nc> <nc>a digital certificate</nc> certifying <nc>a cryptographic key pair</nc> of <nc>a private key</nc> and <nc>a public key</nc> for <nc>the signature authority</nc>; <nc>the signing party computer system</nc> configured for directing <nc>the signature authority</nc> to execute <nc>a</nc> to be signed <nc>electronic document</nc> for creating <nc>a digitally signed electronic document</nc> by generating <nc>a signature creation request</nc> specifying <nc>the</nc> to be signed <nc>electronic document</nc> and communicating <nc>the signature creation request</nc> through <nc>the electronic network</nc> to <nc>the signature authority computer system</nc>; <nc>the signature authority computer system</nc> having <nc>an electronic storage device</nc> <nc>that</nc> stores <nc>the private key</nc> and <nc>the digital certificate</nc> communicated by <nc>the certification authority</nc> through <nc>the electronic network</nc> to <nc>the signature authority computer system</nc> for <nc>use</nc> when constructing <nc>an electronic signature</nc> for indicating <nc>execution</nc> of <nc>the</nc> to be signed <nc>electronic document</nc> as directed by <nc>the signing party</nc> to create <nc>the digitally signed electronic document</nc>; <nc>the signature authority computer system</nc> configured for obtaining in <nc>response</nc> to <nc>a receipt</nc> of <nc>the signature creation request</nc>, <nc>a copy</nc> of <nc>the</nc> to be signed <nc>electronic document</nc> specified in <nc>the signature creation request</nc>; <nc>the signature authority computer system</nc> configured for creating <nc>a signature data structure</nc> <nc>that</nc> includes <nc>an assertion</nc> that <nc>the signature authority</nc> applies <nc>its digital signature</nc> to <nc>the</nc> to be signed <nc>electronic document</nc> for <nc>the purpose</nc> of certifying that <nc>the signing party</nc> has legally signed <nc>the</nc> to be signed <nc>document</nc> as directed in <nc>the signature creation request</nc>, <nc>the signature authority</nc>, creating <nc>the signature data structure</nc> and, with <nc>the retrieved private key and digital certificate</nc>, creating <nc>a digital signing signature</nc> covering <nc>the signature data structure</nc> and the to be signed <nc>document</nc> and resulting in <nc>a digitally signed electronic document</nc>, whereby <nc>a relying party</nc> receiving <nc>the digitally signed electronic document</nc>, relying on <nc>the signature data structure</nc>, the digital signing signature, and <nc>the signature authority digital certificate</nc> for verifying <nc>the digital signing signature</nc> on <nc>the signature data structure</nc> using <nc>the signature authority digital certificate</nc>, to evaluate <nc>a risk</nc> of relying on <nc>the digitally signed electronic document</nc>.
| 4
|
4. <nc>The system</nc> as recited in <nc>claim</nc> 1 , further comprising <nc>a signed document storage database</nc> for storing <nc>the digitally signed electronic document</nc>.
|
10063878
|
15187160
|
1. <nc>An interlayer video decoding method</nc> comprising: obtaining <nc>brightness compensation information</nc> indicating whether <nc>a second layer current block</nc> performs <nc>brightness compensation</nc>; determining whether <nc>a candidate</nc> of <nc>the second layer current block</nc> is usable as <nc>a merge candidate</nc> based on whether <nc>the brightness compensation information</nc> indicates that <nc>the brightness compensation</nc> is performed and whether <nc>the candidate</nc> of <nc>the second layer current block</nc> performs <nc>time direction inter prediction</nc>; generating <nc>a merge candidate list</nc> including <nc>at least one merge candidate</nc> based on <nc>a result</nc> of <nc>the determining</nc>; and determining <nc>motion information</nc> of <nc>the second layer current block</nc> by using <nc>motion information</nc> of one of <nc>the at least one merge candidate</nc>.
| 7
|
7. <nc>The interlayer video decoding method</nc> of <nc>claim</nc> 1 , wherein <nc>the candidate</nc> is based on <nc>motion information</nc> of <nc>a block</nc> corresponding to <nc>a disparity vector</nc> of <nc>the second layer current block</nc> from <nc>a location</nc> of <nc>the second layer current block</nc>, and <nc>the interlayer video decoding method</nc> further comprises, when <nc>the brightness compensation information</nc> indicates that <nc>the brightness compensation</nc> is not performed, determining <nc>a vertical component</nc> of <nc>the disparity vector</nc> to be 0.
|
7646317
|
12110425
|
1. <nc>A decoding method</nc> for mapping <nc>a plurality</nc> of <nc>encoding sequences</nc> to <nc>a plurality</nc> of decoding <nc>sequences</nc>, <nc>each</nc> of <nc>the encoding sequences</nc> including <nc>at least one encoding symbol</nc> chosen from <nc>an encoding symbol set</nc>, <nc>each</nc> of decoding <nc>sequences</nc> including <nc>at least one decoding symbol</nc> chosen from <nc>a decoding symbol set</nc> <nc>which</nc> is used by <nc>non-logographic languages</nc>, <nc>the decoding method</nc> comprising <nc>the steps</nc> of: receiving <nc>an entered encoding symbol</nc>; and combining <nc>the entered encoding symbol</nc> to <nc>an end</nc> of <nc>an input sequence</nc>, wherein <nc>the input sequence</nc> is temporally ambiguous such that <nc>the input sequence</nc> has <nc>possibility</nc> to be interpreted as <nc>at least two different encoding sequence combinations</nc>, <nc>each</nc> of <nc>which</nc> includes at least one of <nc>the encoding sequences</nc>.
| 7
|
7. <nc>The decoding method</nc> of <nc>claim</nc> 1 , further comprises <nc>the step</nc> of determining according to <nc>the mapping</nc> such that <nc>symbols</nc> in <nc>the input sequence</nc> before <nc>the entered encoding symbol</nc> to be <nc>a segmented sequence</nc> including <nc>at least one segment</nc> when <nc>the last two symbols</nc> of <nc>the input sequence</nc> do not form <nc>a part</nc> of <nc>any encoding sequence</nc> in <nc>the mapping</nc>.
|
8161131
|
12058672
|
1. <nc>A method</nc> for delivering <nc>dynamic media content</nc> to <nc>collaborators</nc>, <nc>the method</nc> comprising: providing <nc>collaborative event media content</nc>, wherein <nc>the collaborative event media content</nc> further comprises <nc>a grammar</nc> and <nc>a structured document</nc>, wherein <nc>the grammar</nc> is <nc>a data structure</nc> associating <nc>key phrases</nc> with <nc>presentation actions</nc> <nc>that</nc> facilitates <nc>a collaborator</nc> navigating <nc>the structured document</nc> of <nc>the collaborative event media content</nc> using <nc>speech commands</nc>; providing <nc>data</nc> identifying <nc>a client's location</nc>; storing, in <nc>the context server</nc> in <nc>a data structure</nc> comprising <nc>a dynamic client context</nc> for <nc>the client</nc>, <nc>the data</nc> identifying <nc>the client's location</nc>; detecting <nc>an event</nc> in <nc>dependence</nc> upon <nc>the dynamic client context</nc>, said <nc>event</nc> being characterized by <nc>an event type</nc>; identifying <nc>one or more collaborators</nc> in <nc>dependence</nc> upon <nc>the dynamic client context</nc> and <nc>the event</nc>, <nc>the one or more collaborators</nc> <nc>each</nc> being characterized by <nc>a collaborator classification</nc>; selecting from <nc>the structured document</nc> <nc>a classified structural element</nc> in <nc>dependence</nc> upon <nc>the event type</nc> and <nc>the collaborator classification</nc> for <nc>each</nc> of <nc>the one or more collaborators</nc>; and transmitting <nc>the selected structural element</nc> to <nc>the one or more collaborators</nc>.
| 10
|
10. <nc>The method</nc> of <nc>claim</nc> 1 wherein said <nc>dynamic client context</nc> comprises <nc>at least two data elements</nc>, and said <nc>event</nc> is defined as <nc>a predefined change</nc> in <nc>each</nc> of said <nc>at least two data elements</nc>.
|
7802183
|
10150709
|
1. <nc>An electronic medical record management system</nc> for generating <nc>an electronic medical document</nc> based on <nc>a specific context</nc> from among <nc>a plurality</nc> of <nc>contexts</nc> within <nc>which</nc> <nc>an event</nc> may be documented, <nc>the system</nc> comprising: means for providing <nc>one or more headings</nc> and <nc>a selection</nc> of <nc>available subheadings</nc> corresponding to <nc>each</nc> of <nc>the one or more headings</nc>; means for receiving <nc>requests</nc> from <nc>one or more users</nc> to enter <nc>content</nc> under one or more of <nc>the selection</nc> of <nc>available subheadings</nc> corresponding to at least one of <nc>the one or more headings</nc>; means for converting <nc>each</nc> of <nc>the one or more available subheadings</nc> for <nc>which</nc> <nc>a request</nc> to enter <nc>content</nc> has been received into a corresponding one of <nc>one or more selected subheadings</nc>; means for receiving <nc>requests</nc> from <nc>the one or more users</nc> to associate at least one of <nc>a plurality</nc> of <nc>contexts</nc> with <nc>the one or more selected subheadings</nc> and <nc>headings</nc>; and means for generating <nc>an electronic medical document</nc> including <nc>the entered content</nc> and <nc>the corresponding selected subheadings</nc> and <nc>headings</nc> <nc>that</nc> are associated with <nc>a specific context</nc> from among <nc>the plurality</nc> of <nc>contexts</nc>.
| 2
|
2. <nc>The record management system</nc> of <nc>claim</nc> 1 further comprising <nc>means</nc> for directly entering <nc>quantitative information</nc> into <nc>said electronic document</nc>.
|
9870356
|
14180335
|
1. <nc>A method</nc> implemented by <nc>one or more computing devices</nc>, <nc>the method</nc> comprising: receiving <nc>input data</nc>, <nc>the input data</nc> comprising <nc>linguistic items</nc> including: <nc>a first set</nc> of <nc>linguistic items</nc> with <nc>known intent labels</nc>, <nc>the known intent labels</nc> representing <nc>known relations</nc> between <nc>entities</nc> provided by <nc>a knowledge resource</nc>; and <nc>a second set</nc> of <nc>linguistic items</nc> without <nc>known intent labels</nc> provided by <nc>the knowledge resource</nc>; determining <nc>intents</nc> for <nc>the linguistic items</nc> in <nc>the input data</nc> to produce <nc>intent output information</nc>, the determining comprising: when <nc>a respective linguistic item</nc> corresponds to <nc>a member</nc> of <nc>the first set</nc>, deterministically assigning <nc>a respective known intent</nc> to <nc>the respective linguistic item</nc> based at least on <nc>a respective known intent label</nc> associated with <nc>the respective linguistic item</nc>; and when <nc>the respective linguistic item</nc> corresponds to <nc>a member</nc> of <nc>the second set</nc>, inferring <nc>the intent</nc> associated with <nc>the respective linguistic item</nc> based at least on <nc>selection log data</nc>; and storing <nc>the intent output information</nc> in <nc>a data store</nc>, <nc>the determining</nc> including discovering <nc>a new intent</nc> for <nc>an individual linguistic item</nc> of <nc>the second set</nc> <nc>that</nc> identifies <nc>an individual entity</nc> represented in <nc>the knowledge resource</nc>, <nc>the new intent</nc> identifying <nc>a new relation</nc> for <nc>the individual entity</nc> <nc>that</nc> is not included in <nc>the known relations</nc> provided by <nc>the knowledge resource</nc>, <nc>the selection log data</nc> reflecting <nc>actions</nc> of <nc>users</nc> associated with using <nc>various linguistic items</nc> with <nc>the known intents</nc> and with <nc>the new intent</nc>.
| 14
|
14. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising training <nc>a language understanding model</nc> based at least on <nc>the intent output information</nc>.
|
9064006
|
13592638
|
1. <nc>A method</nc> for providing <nc>natural language query translation</nc>, <nc>the method</nc> comprising: training <nc>a statistical model</nc> according to <nc>a plurality</nc> of <nc>query</nc> click <nc>log data</nc>, <nc>the plurality</nc> of <nc>click</nc> <nc>log data</nc> being mined to train <nc>the statistical model</nc> for <nc>domain detection</nc> in <nc>the absence</nc> of <nc>available in-domain data</nc>; receiving <nc>a natural language query</nc>; translating <nc>the natural language query</nc> into <nc>a search query</nc> according to <nc>the statistical model</nc>; performing <nc>the search query</nc>; and providing <nc>at least one result</nc> associated with performing <nc>the search query</nc>.
| 3
|
3. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the natural language query</nc> is received as <nc>speech</nc>.
|
8078611
|
11648950
|
1. <nc>A method</nc>, comprising: receiving <nc>a query</nc> to <nc>an XML document</nc>, where <nc>the XML document stores</nc> a translation for <nc>a data element</nc>; determining if <nc>the query</nc> includes identifying <nc>elements</nc> <nc>that</nc> identify <nc>the query</nc> as being configured to act on <nc>a translation-enabled database</nc> and, if <nc>the query</nc> includes identifying <nc>elements</nc>, selecting <nc>an explicit query mode</nc> <nc>that</nc> controls <nc>an XPath logic</nc> to use <nc>a normalized translation function</nc> to process <nc>the query</nc>; determining whether <nc>a terminal node</nc> in <nc>an XPath</nc> associated with <nc>the query</nc> refers to <nc>a translatable element</nc>, and if <nc>the terminal node</nc> in <nc>the XPath</nc> associated with <nc>the query</nc> is determined to refer to <nc>a translatable element</nc>, selecting <nc>an implicit query mode</nc> for querying <nc>the XML document</nc>, wherein <nc>the query</nc> is rewritten for <nc>use</nc> by <nc>a translate function</nc>; and querying <nc>the XML document</nc> using <nc>the query</nc> and <nc>the selected query mode</nc>.
| 7
|
7. <nc>The method</nc> of <nc>claim</nc> 1 , where querying <nc>the XML document</nc> using <nc>the explicit query mode</nc> includes controlling <nc>an XPath logic</nc> to use <nc>the normalized translation function</nc>, where <nc>the normalized function</nc> can be operatively connected to <nc>one or more of, a streaming evaluation logic</nc>, and <nc>an XML index logic</nc>.
|
8233726
|
11945978
|
1. <nc>A computer-implemented method</nc> of identifying <nc>a writing system</nc> associated with <nc>a document image</nc> containing <nc>one or more words</nc> written in <nc>the writing system</nc>, <nc>the method</nc> comprising: identifying <nc>a document image fragment</nc> based on <nc>the document image</nc>, wherein <nc>the document image fragment</nc> contains <nc>one or more pixels</nc> from one or more of <nc>the words</nc> in <nc>the document image</nc>; generating <nc>a set</nc> of <nc>sequential features</nc> associated with <nc>the document image fragment</nc>, wherein <nc>each sequential feature</nc> describes <nc>one dimensional graphic information</nc> derived from <nc>the one or more pixels</nc> in <nc>the document image fragment</nc>; identifying <nc>a plurality</nc> of <nc>n</nc><nc>-grams</nc> based on <nc>the set</nc> of <nc>sequential features</nc>, wherein <nc>each n-gram</nc> comprises <nc>an ordered subset</nc> of <nc>sequential features</nc>; generating <nc>a classification score</nc> for <nc>the document image fragment</nc> based at least in <nc>part</nc> on <nc>a frequency</nc> of <nc>occurrence</nc> of <nc>the n</nc><nc>-grams</nc> in <nc>sets</nc> of <nc>sequential features</nc> associated with <nc>known writing systems</nc>, <nc>the classification score</nc> indicating <nc>a likelihood</nc> that <nc>the document image fragment</nc> is written in <nc>the writing system</nc>; and identifying <nc>the writing system</nc> associated with <nc>the document image</nc> based at least in <nc>part</nc> on <nc>the classification score</nc> for <nc>the document image fragment</nc>.
| 27
|
27. <nc>The method</nc> of <nc>claim</nc> 1 , <nc>wherein identification</nc> of <nc>the plurality</nc> of <nc>n</nc><nc>-grams</nc> includes generating <nc>the plurality</nc> of <nc>n</nc><nc>-grams</nc> according to <nc>a sliding window scheme</nc>, wherein <nc>each</nc> generated n-gram in <nc>the plurality</nc> differs from <nc>at least one other n</nc><nc>-gram</nc> in <nc>the plurality</nc> by <nc>one feature</nc>.
|
8122043
|
12494452
|
1. <nc>A method</nc> for ranking <nc>the relevance</nc> of <nc>each</nc> of <nc>a plurality</nc> of <nc>documents</nc> in <nc>a corpus</nc> to <nc>a search query</nc> of <nc>words</nc> comprising <nc>the steps</nc> of: a) grouping <nc>words</nc> in <nc>the search query</nc> by <nc>synonym</nc> into <nc>one or more word groups</nc>, said <nc>grouping</nc> being performed by <nc>a processing unit</nc>; b) for <nc>each word group</nc>, counting <nc>the number</nc> of <nc>instances</nc> (<nc>the “FQ” value</nc>) that <nc>a word</nc> from <nc>the word group</nc> appears in <nc>the search query</nc>, said <nc>counting</nc> being performed by <nc>the processing unit</nc>; c) determining, by <nc>the processing unit</nc>, <nc>the maximum FQ value</nc> among <nc>all the word groups</nc>; d) calculating, by <nc>the processing unit</nc>, <nc>a scaling factor K</nc>; e) for <nc>each word group</nc>, calculating <nc>a term frequency</nc> <nc>(“TF”) value</nc> by dividing <nc>the FQ value</nc> for <nc>the word group</nc> by <nc>the maximum FQ value</nc> and applying <nc>scaling factor K</nc> to <nc>the resulting quotient</nc>, said calculating being performed by <nc>the processing unit</nc>; f) for <nc>each word group</nc>, counting <nc>the number</nc> of <nc>documents</nc> (“<nc>FC</nc>”) in <nc>the corpus</nc> <nc>that</nc> contain <nc>at least one word</nc> from <nc>the word group</nc>, said <nc>counting</nc> being performed by <nc>the processing unit</nc>; <nc>g</nc>) counting <nc>the number</nc> of <nc>documents</nc> (“<nc>N</nc>”) in <nc>the corpus</nc>, said counting being performed by <nc>the processing unit</nc>; h) for <nc>each word group</nc>, calculating <nc>an inverse document frequency</nc> <nc>(“IDF”) value</nc> by dividing <nc>N</nc> by <nc>FC</nc>, adding one to <nc>the resulting quotient</nc>, and taking <nc>the natural logarithm</nc> of <nc>the resulting sum</nc>, said calculating being performed by <nc>the processing unit</nc>; i) for <nc>each word group</nc>, calculating <nc>a TF-IDF value</nc> by multiplying <nc>said TF value</nc> by said <nc>IDF value</nc>, said calculating being performed by <nc>the processing unit</nc>; and <nc>j</nc>) ranking <nc>the relevance</nc> of <nc>each document</nc> in <nc>the corpus</nc> utilizing <nc>the TF-IDF values</nc> for <nc>the word groups</nc> in <nc>the search query</nc>, said <nc>ranking</nc> being performed by <nc>the processing unit</nc>.
| 4
|
4. <nc>The method</nc> of <nc>claim</nc> 1 wherein said <nc>scaling factor K</nc> is <nc>a monotonically decreasing function</nc> over <nc>the domain</nc> of <nc>positive integers</nc> <nc>whose range</nc> does not exceed 1 or fall below 0 where <nc>the domain</nc> represents <nc>the number</nc> of <nc>unique words</nc> (“<nc>C</nc>”) in <nc>the search query</nc>.
|
7536399
|
11166146
|
1. <nc>A data compression method</nc> <nc>that</nc> generates <nc>compressed data</nc> from <nc>a data string</nc> to be compressed, comprising: inputting and retaining, by <nc>an input unit</nc>, <nc>the data string</nc> to be compressed in <nc>an input buffer</nc>; generating and retaining, by <nc>a recent match position list generating unit</nc>, <nc>a recent match position list</nc> having stored therein <nc>a relative position</nc> where <nc>each character</nc> string having <nc>a predetermined length</nc> starting at <nc>each address</nc> in <nc>the input buffer</nc> has most recently appeared; acquiring, by <nc>a repetition candidate acquiring unit</nc>, with <nc>the use</nc> of <nc>the recent match position list</nc>, <nc>a repetition candidate</nc> at <nc>a position</nc> where <nc>a character string</nc> at <nc>a coding position</nc> has previously appeared; comparing, by <nc>a match detecting unit</nc>, a character string starting at <nc>the position</nc> of <nc>the acquired repetition candidate</nc> and <nc>the character string</nc> at <nc>the coding position</nc>, and detecting <nc>a matching character string</nc> from <nc>the position</nc> of <nc>the repetition candidate</nc>; and coding, by <nc>a code generating unit</nc>, <nc>the detected matching character string</nc> having <nc>the longest match length</nc>, wherein in <nc>the candidate acquiring unit</nc>, <nc>a stored value</nc> acquired from <nc>the recent match position list</nc> with <nc>the coding position</nc> being taken as <nc>an address</nc> is taken as <nc>a first candidate</nc> for <nc>a character string repetition position</nc>, in <nc>the match detecting unit</nc>, <nc>a character string</nc> starting at <nc>a position</nc> of <nc>the first candidate</nc> and <nc>the character string</nc> starting at <nc>the coding position</nc> are compared, and <nc>a matching character string</nc> is acquired and coded, <nc>the candidate acquiring unit</nc> further includes: comparing, with <nc>the first candidate</nc> being taken as <nc>an evaluation value</nc>, <nc>a stored value</nc> acquired from <nc>the recent match position list</nc> and <nc>the evaluation value</nc> with <nc>each position</nc> subsequent to <nc>the coding position</nc> being taken as <nc>an address</nc> and, when <nc>the acquired store value</nc> is <nc>a value</nc> previous to <nc>the evaluation value</nc>, acquiring <nc>one or plural subsequent candidates</nc> following <nc>the first candidate</nc> in <nc>order</nc> of <nc>increasing distance</nc> from <nc>the coding position</nc> comparing <nc>each</nc> of <nc>character strings</nc> starting at <nc>the first candidate</nc> and <nc>the subsequent candidates</nc> and <nc>the character string</nc> starting at <nc>the coding position</nc>, and taking, as <nc>a revised first candidate</nc> a character string having <nc>the longest match length</nc> with <nc>respect</nc> to <nc>the character string</nc> at <nc>the coding position</nc>, and taking, as <nc>one or plural subsequent candidates</nc> following <nc>the revised first candidate</nc>, <nc>a stored value</nc> acquired from <nc>the recent match position list</nc> as <nc>revised subsequent candidates</nc> with <nc>a preceding candidate</nc> being taken as <nc>an address</nc>, and in <nc>the match detecting unit</nc>, <nc>each</nc> of <nc>character strings</nc> starting at <nc>the revised first candidate</nc> and <nc>the revised subsequent candidates</nc> and <nc>the character string</nc> starting at <nc>the coding position</nc> are compared, and <nc>a character string</nc> having <nc>the longest match length</nc> is detected and coded.
| 4
|
4. <nc>The data compression method</nc> according to <nc>claim</nc> 1 , wherein <nc>the candidate acquiring unit</nc> further includes: taking <nc>the value</nc> acquired from <nc>the recent match position list</nc> with <nc>the first candidate</nc> being taken as <nc>an address</nc>; and comparing <nc>the stored value</nc> acquired from <nc>the recent match position list</nc> and <nc>the evaluation value</nc> with <nc>each address</nc> subsequent to <nc>the coding position</nc> being taken as <nc>an address</nc> and, when <nc>the acquired stored value</nc> is <nc>a value</nc> previous to <nc>the evaluation value</nc>, acquiring one or <nc>a plurality</nc> of <nc>subsequent candidates</nc> in <nc>order</nc> of <nc>increasing distance</nc> from <nc>the coding position</nc> as, and in <nc>the match detecting unit</nc>, <nc>each</nc> of <nc>character strings</nc> starting at <nc>the first candidate</nc> and <nc>the subsequent candidates</nc> and <nc>the character string</nc> starting at <nc>the coding position</nc> are compared, and <nc>a character string</nc> having <nc>the longest match length</nc> is detected and coded.
|
7630959
|
09947944
|
1. <nc>A method</nc> for processing <nc>a query</nc> on <nc>a database</nc> using <nc>a server</nc> operatively connected to <nc>the database</nc>, <nc>the method</nc> comprising: presenting from <nc>the server</nc>, to <nc>a user</nc>, <nc>a plurality</nc> of <nc>query elements</nc>, <nc>each query element</nc> having <nc>a plurality</nc> of <nc>allowed query element values</nc>, <nc>the plurality</nc> of <nc>allowed query element values</nc> for <nc>each query element</nc> being presented to <nc>the user</nc> for <nc>selection</nc> of <nc>a query element value</nc> from <nc>the plurality</nc> of <nc>allowed query element values</nc>, <nc>the query element values</nc> not selected by <nc>the user</nc> being <nc>non-selected query element values</nc>; receiving at <nc>the server</nc>, from <nc>the user</nc>, <nc>a selected query element value</nc> for at least one of <nc>the plurality</nc> of <nc>query elements</nc>; retrieving <nc>information objects</nc> stored in <nc>the database</nc>, <nc>the database</nc> storing <nc>a first relevance value</nc> <nc>which</nc> defines <nc>a relevance</nc> of <nc>at least one information object</nc> with <nc>respect</nc> to <nc>the selected query element value</nc> and <nc>a second relevance value</nc> <nc>which</nc> defines <nc>a relevance</nc> of <nc>the at least one information object</nc> with <nc>respect</nc> to <nc>the non-selected query element value</nc>, <nc>the first relevance value</nc> and <nc>the second relevance value</nc> being assigned by <nc>a human editor</nc> other than <nc>the user</nc>, <nc>the first relevance value</nc> and <nc>the second relevance value</nc> being assigned before <nc>the plurality</nc> of <nc>query elements</nc> is presented to <nc>the user</nc>, the information objects being retrieved based on <nc>the first relevance value</nc> and <nc>the second relevance value</nc>, <nc>the database</nc> further including <nc>an index</nc>, <nc>the index</nc> associating <nc>at least one information object</nc> stored in <nc>the database</nc> with <nc>at least the selected query element value</nc>; wherein retrieving <nc>the information objects</nc> based on <nc>the first relevance value</nc> and <nc>the second relevance value</nc> is performed by <nc>the server</nc> in <nc>response</nc> to receiving <nc>the selected query element value</nc> for at least one of <nc>the plurality</nc> of <nc>query elements</nc>, and <nc>returning query</nc> results to <nc>the user</nc>, <nc>the query</nc> results corresponding to <nc>the information objects</nc>.
| 6
|
6. <nc>The method</nc> of <nc>claim</nc> 1 , further comprising <nc>mapping</nc> <nc>the retrieved information objects</nc> into <nc>corresponding mapped information objects</nc>.
|
8276101
|
13250089
|
1. <nc>A method</nc> comprising: receiving, using <nc>a presence-sensitive display</nc> coupled to <nc>a computing device</nc>, <nc>a first user input</nc> comprising <nc>a first drawing gesture</nc> associated with <nc>a first area</nc> for <nc>user input</nc> defined at <nc>the presence-sensitive display</nc>, wherein <nc>the first user input</nc> specifies <nc>one or more characters</nc> to be displayed at <nc>the presence-sensitive display</nc>, and wherein <nc>the first drawing gesture</nc> includes <nc>a drawn representation</nc> of <nc>the one or more characters</nc>; receiving, using <nc>the presence-sensitive display</nc>, <nc>a second user input</nc> comprising <nc>a second drawing gesture</nc>, wherein <nc>the second drawing gesture</nc> spans <nc>only the first area</nc> and <nc>a second area</nc> for <nc>user input</nc> defined at <nc>the presence-sensitive display</nc>, and wherein <nc>the second user input</nc> specifies <nc>a first editing operation</nc> associated with <nc>the one or more characters</nc>; applying, by <nc>the computing device</nc>, <nc>the first editing operation</nc> to <nc>the one or more characters</nc> in <nc>response</nc> to receiving <nc>the second user input</nc>; receiving, using <nc>the presence-sensitive display</nc>, <nc>a third user input</nc> comprising <nc>a third drawing gesture</nc>, wherein <nc>the third drawing gesture</nc> spans <nc>the first area</nc>, <nc>the second area</nc>, and <nc>a third area</nc> for <nc>user input</nc> defined at <nc>the presence-sensitive display</nc>, and wherein <nc>the third user input</nc> specifies <nc>a second editing operation</nc> associated with <nc>the one or more characters</nc>; and applying, by <nc>the computing device</nc>, <nc>the second editing operation</nc> to <nc>the one or more characters</nc> in <nc>response</nc> to receiving <nc>the third user input</nc>.
| 16
|
16. <nc>The method</nc> of <nc>claim</nc> 1 , wherein <nc>the first user input</nc>, <nc>the second user input</nc>, and <nc>the third user input</nc> are received at <nc>the presence-sensitive display</nc> without receiving <nc>any input</nc> from an on-<nc>screen</nc> keypad.
|
8027815
|
11751998
|
1. <nc>A non-transitory program storage device</nc> is provided readable by <nc>machine</nc>, tangibly embodying <nc>a program</nc> of <nc>instructions</nc> automatically executable by <nc>the machine</nc> to perform <nc>a method</nc> for determining <nc>a cluster</nc>, <nc>the method</nc> comprising: receiving <nc>input data</nc> on <nc>occurrences</nc> of <nc>a phenomenon</nc> at <nc>locations</nc> and <nc>times</nc> and <nc>data</nc> on <nc>characteristics</nc> of <nc>the locations</nc> and <nc>times</nc>; determining <nc>actual occurrences</nc> of <nc>the phenomenon</nc> at <nc>the three-dimensional space-time points</nc> according to <nc>the input data</nc>; determining <nc>expected occurrences</nc> using <nc>a domain dependent model</nc> for <nc>the phenomenon</nc>, wherein <nc>the expected occurrences</nc> are <nc>a random sampling</nc> of <nc>the data</nc> on <nc>characteristics</nc> of <nc>the locations</nc> and <nc>times</nc>; searching <nc>the expected occurrences</nc> for <nc>a first plurality</nc> of <nc>candidate solutions</nc> and <nc>the actual occurrences</nc> for <nc>a second plurality</nc> of <nc>candidate solutions</nc>; determining <nc>a strength</nc> of <nc>each</nc> of <nc>the first and second plurality</nc> of <nc>candidate solutions</nc>; selecting <nc>a convex container shape</nc> from <nc>the second plurality</nc> of <nc>candidate solutions</nc> having <nc>a strength</nc> unlikely to exist in <nc>the expected occurrences</nc> for determining <nc>a cluster</nc> in <nc>the input data</nc>; and determining <nc>a solution</nc> represented as <nc>a set</nc> of <nc>points</nc> <nc>that</nc> conforms to <nc>a volume</nc> of <nc>the selected convex container shape</nc> for <nc>a cluster</nc>.
| 2
|
2. <nc>The program storage device</nc> of <nc>claim</nc> 1 , wherein <nc>the method</nc> further comprises <nc>caching solutions</nc> conforming to <nc>the selected convex container shape</nc>.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.