data/subtitles/Artificial Intelligence.vtt

WEBVTT - Subtitles by: DownloadYoutubeSubtitles.com

00:00:02.590 --> 00:00:12.400
[Music]

00:00:10.160 --> 00:00:16.200
introduction of artificial intelligence

00:00:12.400 --> 00:00:16.200
and machine learning

00:00:16.640 --> 00:00:21.680
by the end of this lesson you will be

00:00:18.480 --> 00:00:23.519
able to define artificial intelligence

00:00:21.680 --> 00:00:26.880
describe the relationship between

00:00:23.519 --> 00:00:29.439
artificial intelligence and data science

00:00:26.880 --> 00:00:31.119
define machine learning

00:00:29.439 --> 00:00:33.440
describe the relationship between

00:00:31.119 --> 00:00:35.760
machine learning artificial intelligence

00:00:33.440 --> 00:00:37.760
and data science

00:00:35.760 --> 00:00:38.960
describe different machine learning

00:00:37.760 --> 00:00:41.200
approaches

00:00:38.960 --> 00:00:43.360
identify the applications of machine

00:00:41.200 --> 00:00:45.280
learning

00:00:43.360 --> 00:00:48.480
let's understand how the field of

00:00:45.280 --> 00:00:50.640
artificial intelligence emerged

00:00:48.480 --> 00:00:52.640
let's first understand the reason behind

00:00:50.640 --> 00:00:54.719
the emergence of a.i

00:00:52.640 --> 00:00:57.039
data economy is one of the factors

00:00:54.719 --> 00:00:59.359
behind the emergence of ai

00:00:57.039 --> 00:01:01.920
it refers to how much data has grown

00:00:59.359 --> 00:01:04.239
over the past few years and how much

00:01:01.920 --> 00:01:06.159
more it can grow in the coming years

00:01:04.239 --> 00:01:08.320
when you look at this graph you can

00:01:06.159 --> 00:01:10.000
clearly understand how the volume of

00:01:08.320 --> 00:01:12.720
data has grown

00:01:10.000 --> 00:01:15.840
you can see that since 2009 the data

00:01:12.720 --> 00:01:18.000
volume has increased by 44 times with

00:01:15.840 --> 00:01:20.560
the help of social websites

00:01:18.000 --> 00:01:23.119
the explosion of data has given rise to

00:01:20.560 --> 00:01:25.360
a new economy and there is a constant

00:01:23.119 --> 00:01:29.040
battle for ownership of data between

00:01:25.360 --> 00:01:31.119
companies to derive benefits from it

00:01:29.040 --> 00:01:33.680
now that you know that data has grown at

00:01:31.119 --> 00:01:36.000
a rapid pace in the past few years and

00:01:33.680 --> 00:01:38.560
is going to continue to grow

00:01:36.000 --> 00:01:40.960
let's understand the need for ai

00:01:38.560 --> 00:01:44.000
as you know the increase in data volume

00:01:40.960 --> 00:01:46.320
has given rise to big data which helps

00:01:44.000 --> 00:01:49.520
manage huge amounts of data

00:01:46.320 --> 00:01:52.000
data science helps analyze that data so

00:01:49.520 --> 00:01:54.640
the science associated with data is

00:01:52.000 --> 00:01:56.720
going toward a new paradigm

00:01:54.640 --> 00:01:59.600
where one can teach machines to learn

00:01:56.720 --> 00:02:01.840
from data and drive a variety of useful

00:01:59.600 --> 00:02:04.159
insights giving rise to artificial

00:02:01.840 --> 00:02:06.840
intelligence

00:02:04.159 --> 00:02:09.360
now you may ask what is artificial

00:02:06.840 --> 00:02:11.680
intelligence artificial intelligence

00:02:09.360 --> 00:02:14.640
refers to the intelligence displayed by

00:02:11.680 --> 00:02:15.840
machines that simulates human and animal

00:02:14.640 --> 00:02:18.319
intelligence

00:02:15.840 --> 00:02:20.400
it involves intelligence agents

00:02:18.319 --> 00:02:23.120
the autonomous entities that perceive

00:02:20.400 --> 00:02:25.360
their environment and take actions that

00:02:23.120 --> 00:02:26.800
maximize their chances of success at a

00:02:25.360 --> 00:02:28.959
given goal

00:02:26.800 --> 00:02:31.280
artificial intelligence is a technique

00:02:28.959 --> 00:02:33.840
that enables computers to mimic human

00:02:31.280 --> 00:02:36.560
intelligence using logic

00:02:33.840 --> 00:02:38.400
it is a program that can sense reason

00:02:36.560 --> 00:02:40.319
and act

00:02:38.400 --> 00:02:43.200
let's look at some of the areas where

00:02:40.319 --> 00:02:45.680
artificial intelligence is used

00:02:43.200 --> 00:02:47.519
artificial intelligence is redefining

00:02:45.680 --> 00:02:50.480
industries by providing greater

00:02:47.519 --> 00:02:51.760
personalization to users and automating

00:02:50.480 --> 00:02:54.000
processes

00:02:51.760 --> 00:02:57.519
one example of artificial intelligence

00:02:54.000 --> 00:02:59.360
in practice is self-driving cars

00:02:57.519 --> 00:03:02.400
self-driving cars are computer

00:02:59.360 --> 00:03:04.640
controlled cars that drive themselves

00:03:02.400 --> 00:03:06.800
in these cars human drivers are never

00:03:04.640 --> 00:03:08.480
required to take control to safely

00:03:06.800 --> 00:03:11.120
operate the vehicle

00:03:08.480 --> 00:03:13.599
these cars are also known as autonomous

00:03:11.120 --> 00:03:16.560
or driverless cars

00:03:13.599 --> 00:03:19.200
let's see how apple uses ai

00:03:16.560 --> 00:03:20.879
iphone users can experience the power of

00:03:19.200 --> 00:03:23.040
siri the voice

00:03:20.879 --> 00:03:24.800
it simplifies navigating through your

00:03:23.040 --> 00:03:27.440
iphone as it listens to your voice

00:03:24.800 --> 00:03:29.599
commands to perform tasks

00:03:27.440 --> 00:03:32.879
for instance you can ask siri to call

00:03:29.599 --> 00:03:36.080
your friend or to play music siri is fun

00:03:32.879 --> 00:03:38.480
and is extremely convenient to use

00:03:36.080 --> 00:03:40.560
another example is google's alphago

00:03:38.480 --> 00:03:42.480
which is a computer program that plays

00:03:40.560 --> 00:03:44.879
the board game go

00:03:42.480 --> 00:03:47.599
it is the first computer program to

00:03:44.879 --> 00:03:49.840
defeat a world champion at the ancient

00:03:47.599 --> 00:03:52.560
chinese game of go

00:03:49.840 --> 00:03:54.640
amazon echo is another product it's a

00:03:52.560 --> 00:03:57.120
home control chatbot device that

00:03:54.640 --> 00:03:59.680
responds to humans according to what

00:03:57.120 --> 00:04:02.080
they are saying it responds by playing

00:03:59.680 --> 00:04:04.080
music movies and more

00:04:02.080 --> 00:04:06.239
if you've got compatible smart home

00:04:04.080 --> 00:04:09.120
devices you can tell echo to dim the

00:04:06.239 --> 00:04:11.439
lights or turn appliances on or off you

00:04:09.120 --> 00:04:14.879
can use ai and chess and here is an

00:04:11.439 --> 00:04:16.959
example of a concierge robot from ibm

00:04:14.879 --> 00:04:20.079
called ibm watson

00:04:16.959 --> 00:04:22.320
the ibm watson ai has typically been in

00:04:20.079 --> 00:04:26.000
the headlines for composing music

00:04:22.320 --> 00:04:28.320
playing chess and even cooking food

00:04:26.000 --> 00:04:30.560
let's move ahead and look at some sci-fi

00:04:28.320 --> 00:04:32.000
movies with the concept of artificial

00:04:30.560 --> 00:04:34.320
intelligence

00:04:32.000 --> 00:04:36.800
the films featuring ai reflect the

00:04:34.320 --> 00:04:39.520
ever-changing spectrum of our emotions

00:04:36.800 --> 00:04:41.840
regarding the machines we have created

00:04:39.520 --> 00:04:44.080
humans are fascinated by the concept of

00:04:41.840 --> 00:04:46.639
artificial intelligence and this is

00:04:44.080 --> 00:04:48.240
reflected in the wide range of movies on

00:04:46.639 --> 00:04:50.800
ai

00:04:48.240 --> 00:04:53.360
recommendations systems are used by a

00:04:50.800 --> 00:04:56.160
lot of e-commerce companies let's see

00:04:53.360 --> 00:04:58.560
how they work

00:04:56.160 --> 00:05:00.800
amazon collects data from users and

00:04:58.560 --> 00:05:03.759
recommends the best product according to

00:05:00.800 --> 00:05:05.520
the user's buying or shopping pattern

00:05:03.759 --> 00:05:08.720
for example when you search for a

00:05:05.520 --> 00:05:10.400
specific product in the amazon store and

00:05:08.720 --> 00:05:12.800
add it to your cart

00:05:10.400 --> 00:05:14.479
amazon recommends some relevant products

00:05:12.800 --> 00:05:16.160
based on your past shopping and

00:05:14.479 --> 00:05:18.479
searching pattern

00:05:16.160 --> 00:05:20.400
so before you buy the selected product

00:05:18.479 --> 00:05:22.960
you get recommendations based on your

00:05:20.400 --> 00:05:25.199
interest and there is a possibility that

00:05:22.960 --> 00:05:28.240
you may also buy the relevant product

00:05:25.199 --> 00:05:30.000
with a selected product if not you have

00:05:28.240 --> 00:05:33.280
the chance to compare the selected

00:05:30.000 --> 00:05:35.360
product with the recommended products

00:05:33.280 --> 00:05:37.039
now let's move ahead and understand the

00:05:35.360 --> 00:05:39.840
relationship between artificial

00:05:37.039 --> 00:05:42.000
intelligence machine learning and data

00:05:39.840 --> 00:05:43.680
science

00:05:42.000 --> 00:05:46.560
even though the terms artificial

00:05:43.680 --> 00:05:49.520
intelligence ai machine learning and

00:05:46.560 --> 00:05:51.360
data science fall in the same domain and

00:05:49.520 --> 00:05:54.160
are connected to each other they have

00:05:51.360 --> 00:05:56.320
their specific applications and meaning

00:05:54.160 --> 00:05:58.000
let's try to understand a little about

00:05:56.320 --> 00:06:00.479
each of these terms

00:05:58.000 --> 00:06:02.800
artificial intelligence systems mimic or

00:06:00.479 --> 00:06:04.880
replicate human intelligence

00:06:02.800 --> 00:06:07.440
machine learning provides systems the

00:06:04.880 --> 00:06:09.440
ability to automatically learn and

00:06:07.440 --> 00:06:12.000
improve from the experiences without

00:06:09.440 --> 00:06:14.160
being explicitly programmed

00:06:12.000 --> 00:06:17.360
data science is an umbrella term that

00:06:14.160 --> 00:06:20.240
encompasses data analytics data mining

00:06:17.360 --> 00:06:22.880
machine learning artificial intelligence

00:06:20.240 --> 00:06:25.120
and several other related disciplines

00:06:22.880 --> 00:06:26.960
let's look at the flow diagram and try

00:06:25.120 --> 00:06:27.919
to understand the relationship between

00:06:26.960 --> 00:06:30.800
ai

00:06:27.919 --> 00:06:33.680
machine learning and data science

00:06:30.800 --> 00:06:35.680
interestingly ml is also an element of

00:06:33.680 --> 00:06:38.160
artificial intelligence

00:06:35.680 --> 00:06:40.000
so the first step is data gathering and

00:06:38.160 --> 00:06:42.319
data transformation

00:06:40.000 --> 00:06:43.520
this step basically comes under data

00:06:42.319 --> 00:06:45.600
science

00:06:43.520 --> 00:06:47.759
data transformation is the process of

00:06:45.600 --> 00:06:50.080
converting data from one format or

00:06:47.759 --> 00:06:51.199
structure into another format or

00:06:50.080 --> 00:06:53.199
structure

00:06:51.199 --> 00:06:55.759
data transformation is important to

00:06:53.199 --> 00:06:57.759
activities such as data management and

00:06:55.759 --> 00:06:59.759
data integration

00:06:57.759 --> 00:07:02.000
after gathering data we would want to

00:06:59.759 --> 00:07:04.319
use the data to make predictions and

00:07:02.000 --> 00:07:06.880
derive insights in order to get

00:07:04.319 --> 00:07:08.639
predictions out of the data set we use

00:07:06.880 --> 00:07:11.599
machine learning techniques such as

00:07:08.639 --> 00:07:14.800
supervised learning or unsupervised

00:07:11.599 --> 00:07:16.639
learning on an overview level supervised

00:07:14.800 --> 00:07:18.960
and unsupervised learning are the

00:07:16.639 --> 00:07:21.199
machine learning techniques used to

00:07:18.960 --> 00:07:22.400
extract predictions from a given data

00:07:21.199 --> 00:07:24.880
set

00:07:22.400 --> 00:07:27.199
now you must be thinking where deep

00:07:24.880 --> 00:07:29.840
learning comes into the picture

00:07:27.199 --> 00:07:32.800
deep learning is a subfield of machine

00:07:29.840 --> 00:07:35.680
learning involved with algorithms

00:07:32.800 --> 00:07:37.199
it uses artificial neural networks which

00:07:35.680 --> 00:07:39.199
are modeled on the structure and

00:07:37.199 --> 00:07:40.560
performance of neurons in the human

00:07:39.199 --> 00:07:42.880
brain

00:07:40.560 --> 00:07:45.199
deep learning is most effective when

00:07:42.880 --> 00:07:46.160
there isn't a clear structure to the

00:07:45.199 --> 00:07:48.160
data

00:07:46.160 --> 00:07:49.680
that you can just exploit and build

00:07:48.160 --> 00:07:52.240
features around

00:07:49.680 --> 00:07:54.240
now the next step in the flow diagram is

00:07:52.240 --> 00:07:55.199
to get insights from predictions being

00:07:54.240 --> 00:07:57.680
made

00:07:55.199 --> 00:08:00.319
in order to do so you need to use data

00:07:57.680 --> 00:08:02.400
analysis which actually is the process

00:08:00.319 --> 00:08:04.080
under data science

00:08:02.400 --> 00:08:06.240
now when you are done with all of these

00:08:04.080 --> 00:08:07.280
you must want your data to perform some

00:08:06.240 --> 00:08:10.160
actions

00:08:07.280 --> 00:08:12.080
this is where ai comes into the picture

00:08:10.160 --> 00:08:14.479
artificial intelligence combines

00:08:12.080 --> 00:08:17.039
predictions and insights to perform

00:08:14.479 --> 00:08:19.680
actions based on the human decision and

00:08:17.039 --> 00:08:21.759
automated decision

00:08:19.680 --> 00:08:23.440
now let's move ahead and understand the

00:08:21.759 --> 00:08:26.240
relationship between artificial

00:08:23.440 --> 00:08:28.160
intelligence machine learning and data

00:08:26.240 --> 00:08:30.000
science

00:08:28.160 --> 00:08:32.080
let's look at the relationship between

00:08:30.000 --> 00:08:33.200
artificial intelligence and machine

00:08:32.080 --> 00:08:35.120
learning

00:08:33.200 --> 00:08:36.959
artificial intelligence is the

00:08:35.120 --> 00:08:39.039
engineering of making intelligent

00:08:36.959 --> 00:08:41.279
machines and programs

00:08:39.039 --> 00:08:44.080
machine learning provides systems the

00:08:41.279 --> 00:08:46.959
ability to learn from past experiences

00:08:44.080 --> 00:08:49.360
without being explicitly programmed

00:08:46.959 --> 00:08:51.839
machine learning allows machines to gain

00:08:49.360 --> 00:08:54.839
intelligence thereby enabling artificial

00:08:51.839 --> 00:08:54.839
intelligence

00:08:54.959 --> 00:08:58.880
let's now understand the relationship

00:08:56.800 --> 00:09:00.000
between machine learning and data

00:08:58.880 --> 00:09:02.160
science

00:09:00.000 --> 00:09:03.600
data science and machine learning go

00:09:02.160 --> 00:09:06.320
hand in hand

00:09:03.600 --> 00:09:08.640
data science helps evaluate data for

00:09:06.320 --> 00:09:10.640
machine learning algorithms

00:09:08.640 --> 00:09:13.040
data science covers the whole spectrum

00:09:10.640 --> 00:09:14.880
of data processing while machine

00:09:13.040 --> 00:09:18.240
learning has the algorithmic or

00:09:14.880 --> 00:09:18.240
statistical aspects

00:09:18.640 --> 00:09:23.839
data science is the use of statistical

00:09:21.040 --> 00:09:26.080
methods to find patterns in the data

00:09:23.839 --> 00:09:28.720
statistical machine learning uses the

00:09:26.080 --> 00:09:31.120
same techniques as data science

00:09:28.720 --> 00:09:34.000
data science includes various techniques

00:09:31.120 --> 00:09:37.040
like statistical modeling visualization

00:09:34.000 --> 00:09:39.440
and pattern recognition machine learning

00:09:37.040 --> 00:09:44.080
focuses on developing algorithms from

00:09:39.440 --> 00:09:47.680
the data provided by making predictions

00:09:44.080 --> 00:09:49.920
so what is machine learning

00:09:47.680 --> 00:09:52.560
machine learning is the capability of an

00:09:49.920 --> 00:09:55.360
artificial intelligence system to learn

00:09:52.560 --> 00:09:57.600
by extracting patterns from data

00:09:55.360 --> 00:09:59.600
it usually delivers quicker more

00:09:57.600 --> 00:10:02.480
accurate results to help you spot

00:09:59.600 --> 00:10:04.399
profitable opportunities or dangerous

00:10:02.480 --> 00:10:06.399
risks

00:10:04.399 --> 00:10:09.040
now you must be curious to understand

00:10:06.399 --> 00:10:11.279
the features of machine learning machine

00:10:09.040 --> 00:10:14.000
learning uses the data to detect

00:10:11.279 --> 00:10:16.480
patterns in a data set and adjust

00:10:14.000 --> 00:10:18.720
program actions accordingly

00:10:16.480 --> 00:10:20.640
pattern detection can be defined as the

00:10:18.720 --> 00:10:23.200
classification of data based on

00:10:20.640 --> 00:10:25.360
knowledge already gained or on

00:10:23.200 --> 00:10:26.800
statistical information extracted from

00:10:25.360 --> 00:10:28.640
the patterns

00:10:26.800 --> 00:10:30.480
it focuses on the development of

00:10:28.640 --> 00:10:32.480
computer programs that can teach

00:10:30.480 --> 00:10:34.560
themselves to grow and change

00:10:32.480 --> 00:10:37.279
when exposed to new data by using a

00:10:34.560 --> 00:10:39.760
method called reinforcement learning

00:10:37.279 --> 00:10:42.399
it uses external feedback to teach the

00:10:39.760 --> 00:10:44.880
system to change its internal workings

00:10:42.399 --> 00:10:46.880
in order to guess better next time

00:10:44.880 --> 00:10:49.600
it enables computers to find hidden

00:10:46.880 --> 00:10:52.640
insights using iterative algorithms

00:10:49.600 --> 00:10:55.120
without being explicitly programmed

00:10:52.640 --> 00:10:57.519
machine learning uses algorithms that

00:10:55.120 --> 00:11:00.399
learn from previous data to help produce

00:10:57.519 --> 00:11:02.640
reliable and repeatable decisions it

00:11:00.399 --> 00:11:04.560
automates analytical model building

00:11:02.640 --> 00:11:07.360
using the statistical and machine

00:11:04.560 --> 00:11:10.240
learning algorithms that tease patterns

00:11:07.360 --> 00:11:13.200
and relationships from data and express

00:11:10.240 --> 00:11:15.120
them as mathematical equations

00:11:13.200 --> 00:11:18.160
let's understand the different machine

00:11:15.120 --> 00:11:18.160
learning approaches

00:11:18.880 --> 00:11:23.519
so what is the actual difference between

00:11:21.519 --> 00:11:26.560
traditional programming and machine

00:11:23.519 --> 00:11:27.360
learning in traditional programming data

00:11:26.560 --> 00:11:30.320
and

00:11:27.360 --> 00:11:32.720
is provided to the computer it processes

00:11:30.320 --> 00:11:34.560
them and gives the output however the

00:11:32.720 --> 00:11:37.360
machine learning approach is very

00:11:34.560 --> 00:11:40.959
different in machine learning algorithms

00:11:37.360 --> 00:11:43.360
are applied on the given data and output

00:11:40.959 --> 00:11:46.000
the result of the applied algorithm and

00:11:43.360 --> 00:11:49.360
calculations is a learning model that

00:11:46.000 --> 00:11:51.680
helps machine to learn from the data

00:11:49.360 --> 00:11:54.320
in traditional programming you code the

00:11:51.680 --> 00:11:56.560
behavior of the program but in machine

00:11:54.320 --> 00:11:59.120
learning you leave a lot of that to the

00:11:56.560 --> 00:12:00.560
machine to learn from data

00:11:59.120 --> 00:12:03.040
now let's first understand the

00:12:00.560 --> 00:12:05.040
traditional programming approach

00:12:03.040 --> 00:12:07.680
traditionally you would hard code the

00:12:05.040 --> 00:12:10.320
decision rules for a problem at hand

00:12:07.680 --> 00:12:12.240
evaluate the results of the program and

00:12:10.320 --> 00:12:15.279
if the results were satisfactory the

00:12:12.240 --> 00:12:17.680
program would be deployed in production

00:12:15.279 --> 00:12:20.079
if the results were not as expected one

00:12:17.680 --> 00:12:22.720
would review the errors change the

00:12:20.079 --> 00:12:25.279
program and evaluate it again

00:12:22.720 --> 00:12:28.800
this iterative process continues till

00:12:25.279 --> 00:12:31.200
one gets the expected result

00:12:28.800 --> 00:12:33.120
what is the machine learning approach in

00:12:31.200 --> 00:12:35.920
the new machine learning approach the

00:12:33.120 --> 00:12:38.240
decision rules are not hard coded the

00:12:35.920 --> 00:12:40.160
problem is solved by training a model

00:12:38.240 --> 00:12:43.279
with the training data in order to

00:12:40.160 --> 00:12:45.760
derive or learn an algorithm that best

00:12:43.279 --> 00:12:48.639
represents the relationship between the

00:12:45.760 --> 00:12:51.680
input and the output this trained model

00:12:48.639 --> 00:12:53.839
is then evaluated against test data if

00:12:51.680 --> 00:12:56.160
the results were satisfactory the model

00:12:53.839 --> 00:12:58.160
would be deployed in production and if

00:12:56.160 --> 00:13:01.920
the results are not satisfactory the

00:12:58.160 --> 00:13:05.360
training is repeated with some changes

00:13:01.920 --> 00:13:05.360
machine learning techniques

00:13:05.600 --> 00:13:09.360
machine learning uses a number of

00:13:07.600 --> 00:13:12.079
theories and techniques from data

00:13:09.360 --> 00:13:14.839
science here are some machine learning

00:13:12.079 --> 00:13:18.480
techniques classification

00:13:14.839 --> 00:13:21.440
categorization clustering trend analysis

00:13:18.480 --> 00:13:22.959
anomaly detection visualization and

00:13:21.440 --> 00:13:26.079
decision making

00:13:22.959 --> 00:13:28.160
let's look at these techniques

00:13:26.079 --> 00:13:30.160
classification is a technique in which

00:13:28.160 --> 00:13:33.360
the computer program learns from the

00:13:30.160 --> 00:13:35.040
data input given to it and then uses

00:13:33.360 --> 00:13:36.639
this learning to classify new

00:13:35.040 --> 00:13:38.959
observations

00:13:36.639 --> 00:13:41.839
classification is used for predicting

00:13:38.959 --> 00:13:44.000
discrete responses classification is

00:13:41.839 --> 00:13:48.360
used when we are training a model to

00:13:44.000 --> 00:13:50.399
predict qualitative targets

00:13:48.360 --> 00:13:52.959
categorization is a technique of

00:13:50.399 --> 00:13:55.440
organizing data into categories for its

00:13:52.959 --> 00:13:57.839
most effective and efficient use

00:13:55.440 --> 00:14:00.959
it makes free text searches faster and

00:13:57.839 --> 00:14:03.279
provides a better user experience

00:14:00.959 --> 00:14:05.120
clustering is a technique of grouping a

00:14:03.279 --> 00:14:07.199
set of objects in such a way that

00:14:05.120 --> 00:14:09.600
objects in the same group are most

00:14:07.199 --> 00:14:10.880
similar to each other than to those in

00:14:09.600 --> 00:14:13.120
other groups

00:14:10.880 --> 00:14:14.959
it is basically a collection of objects

00:14:13.120 --> 00:14:18.000
on the basis of similarity and

00:14:14.959 --> 00:14:20.320
dissimilarity between them

00:14:18.000 --> 00:14:22.240
trend analysis is a technique aimed at

00:14:20.320 --> 00:14:24.639
projecting both current and future

00:14:22.240 --> 00:14:27.279
movement of events through the use of

00:14:24.639 --> 00:14:29.440
time series data analysis

00:14:27.279 --> 00:14:32.399
it represents variations of low

00:14:29.440 --> 00:14:36.399
frequency in a time series the high and

00:14:32.399 --> 00:14:38.880
medium frequency fluctuations being out

00:14:36.399 --> 00:14:41.519
anomaly detection is a technique to

00:14:38.880 --> 00:14:44.560
identify cases that are unusual within

00:14:41.519 --> 00:14:46.639
data that is seemingly homogenous

00:14:44.560 --> 00:14:48.720
anomaly detection can be a key for

00:14:46.639 --> 00:14:51.360
solving intrusions by indicating a

00:14:48.720 --> 00:14:55.639
presence of intended or unintended

00:14:51.360 --> 00:14:58.000
induced attacks defects faults and so on

00:14:55.639 --> 00:15:01.680
visualization is a technique to present

00:14:58.000 --> 00:15:03.760
data in a pictorial or graphical format

00:15:01.680 --> 00:15:06.320
it enables decision makers to see

00:15:03.760 --> 00:15:08.399
analytics presented visually

00:15:06.320 --> 00:15:10.880
when data is shown in the form of

00:15:08.399 --> 00:15:12.480
pictures it becomes easy for users to

00:15:10.880 --> 00:15:14.639
understand it

00:15:12.480 --> 00:15:16.560
decision making is a technique or skill

00:15:14.639 --> 00:15:19.600
that provides you with the ability to

00:15:16.560 --> 00:15:22.800
influence managerial decisions with data

00:15:19.600 --> 00:15:24.720
as evidence for those possibilities

00:15:22.800 --> 00:15:26.800
now i am sure you have a better

00:15:24.720 --> 00:15:29.120
understanding of the overview of machine

00:15:26.800 --> 00:15:31.680
learning so let's look at some real-time

00:15:29.120 --> 00:15:33.839
applications of machine learning

00:15:31.680 --> 00:15:36.160
artificial intelligence and machine

00:15:33.839 --> 00:15:38.320
learning are being increasingly used in

00:15:36.160 --> 00:15:40.399
various functions such as image

00:15:38.320 --> 00:15:44.079
processing robotics

00:15:40.399 --> 00:15:46.399
data mining video games text analysis

00:15:44.079 --> 00:15:48.320
and healthcare let's look at each of

00:15:46.399 --> 00:15:51.199
them in more details

00:15:48.320 --> 00:15:53.279
so what is image processing it is a

00:15:51.199 --> 00:15:55.519
technique to convert an image into a

00:15:53.279 --> 00:15:58.160
digital format and perform some

00:15:55.519 --> 00:16:00.560
operations on it so as to induce an

00:15:58.160 --> 00:16:02.800
enhanced image or to extract some

00:16:00.560 --> 00:16:04.720
helpful information from it

00:16:02.800 --> 00:16:06.399
let's look at some of the examples of

00:16:04.720 --> 00:16:08.880
image processing

00:16:06.399 --> 00:16:10.959
facebook does automatic face tagging by

00:16:08.880 --> 00:16:14.000
recognizing a face from a previous

00:16:10.959 --> 00:16:15.839
user's tagged photos another example is

00:16:14.000 --> 00:16:19.279
optional character recognition which

00:16:15.839 --> 00:16:21.040
scans printed docs to digitize the text

00:16:19.279 --> 00:16:23.360
self-driving cars are another big

00:16:21.040 --> 00:16:26.079
example of image processing

00:16:23.360 --> 00:16:27.839
autopilot is an optional drive system

00:16:26.079 --> 00:16:30.399
for tesla cars

00:16:27.839 --> 00:16:34.160
when autopilot is engaged cars can

00:16:30.399 --> 00:16:37.680
self-steer adjust speed detect nearby

00:16:34.160 --> 00:16:40.079
obstacles apply the brakes and park

00:16:37.680 --> 00:16:41.040
now let's see how robotics uses machine

00:16:40.079 --> 00:16:43.120
learning

00:16:41.040 --> 00:16:44.720
robots are machines that can be used to

00:16:43.120 --> 00:16:47.199
do certain jobs

00:16:44.720 --> 00:16:49.199
some of the examples of robotics are

00:16:47.199 --> 00:16:52.079
where a humanoid robot can read the

00:16:49.199 --> 00:16:54.079
emotions of human beings or

00:16:52.079 --> 00:16:57.040
an industrial robot is used for

00:16:54.079 --> 00:16:58.880
assembling and manufacturing products

00:16:57.040 --> 00:17:01.360
so let's look at some real-time

00:16:58.880 --> 00:17:04.880
applications of machine learning

00:17:01.360 --> 00:17:07.199
let's see what data mining is it is the

00:17:04.880 --> 00:17:08.160
method of analyzing hidden patterns in

00:17:07.199 --> 00:17:10.160
data

00:17:08.160 --> 00:17:11.679
let's look at some of the applications

00:17:10.160 --> 00:17:13.919
of data mining

00:17:11.679 --> 00:17:16.000
it is used for anomaly detection to

00:17:13.919 --> 00:17:18.720
detect credit card fraud and to

00:17:16.000 --> 00:17:21.839
determine which transactions vary from

00:17:18.720 --> 00:17:24.160
usual purchasing patterns

00:17:21.839 --> 00:17:26.799
it is also used in market basket

00:17:24.160 --> 00:17:30.559
analysis which is used to detect which

00:17:26.799 --> 00:17:30.559
items are often bought together

00:17:31.679 --> 00:17:38.720
it can be used for grouping where it

00:17:33.679 --> 00:17:38.720
classifies users based on their profiles

00:17:38.799 --> 00:17:43.280
machine learning is also applied in many

00:17:41.039 --> 00:17:46.799
video games in order to give predictions

00:17:43.280 --> 00:17:48.640
based on data in a pokemon go battle

00:17:46.799 --> 00:17:50.480
there is a lot of data to take into

00:17:48.640 --> 00:17:51.760
account to correctly predict the winner

00:17:50.480 --> 00:17:53.600
of a battle

00:17:51.760 --> 00:17:56.000
and this is where machine learning

00:17:53.600 --> 00:17:58.400
becomes useful a machine learning

00:17:56.000 --> 00:18:01.280
classifier will predict the result of

00:17:58.400 --> 00:18:03.520
the match based on this data

00:18:01.280 --> 00:18:05.440
let's move on to one of the most popular

00:18:03.520 --> 00:18:07.360
applications of machine learning which

00:18:05.440 --> 00:18:09.919
is text analysis

00:18:07.360 --> 00:18:11.840
it is the automated process of obtaining

00:18:09.919 --> 00:18:14.640
information from text

00:18:11.840 --> 00:18:17.600
one example of text analysis is spam

00:18:14.640 --> 00:18:19.039
filtering which is used to detect spam

00:18:17.600 --> 00:18:21.440
in emails

00:18:19.039 --> 00:18:24.160
another example is sentimental analysis

00:18:21.440 --> 00:18:26.400
which is used for classifying an opinion

00:18:24.160 --> 00:18:28.799
as positive negative or neutral it

00:18:26.400 --> 00:18:31.360
detects public sentiment in twitter feed

00:18:28.799 --> 00:18:33.280
or filters customer complaints

00:18:31.360 --> 00:18:36.320
it is also used for information

00:18:33.280 --> 00:18:40.880
extraction such as extracting specific

00:18:36.320 --> 00:18:40.880
data address keyword or entities

00:18:41.200 --> 00:18:45.520
there are many applications of machine

00:18:43.280 --> 00:18:48.720
learning in the healthcare industry

00:18:45.520 --> 00:18:51.600
identifying disease and diagnosis

00:18:48.720 --> 00:18:54.400
drug discovery and manufacturing medical

00:18:51.600 --> 00:18:56.320
imaging diagnosis and so on

00:18:54.400 --> 00:18:58.480
some of the companies that use machine

00:18:56.320 --> 00:19:01.200
learning have revolutionized the health

00:18:58.480 --> 00:19:02.160
care industry are google deep mind

00:19:01.200 --> 00:19:06.320
health

00:19:02.160 --> 00:19:08.430
bio beats health fidelity and ginger dot

00:19:06.320 --> 00:19:12.880
io

00:19:08.430 --> 00:19:14.960
[Music]

00:19:12.880 --> 00:19:14.960
you