rmtariq commited on
Commit
473076a
·
verified ·
1 Parent(s): 2ea9ba2

Update classification_rules.txt

Browse files
Files changed (1) hide show
  1. classification_rules.txt +57 -1
classification_rules.txt CHANGED
@@ -6,7 +6,63 @@ GOVERNMENT: kerajaan,menteri,perdana menteri,anwar ibrahim,anwar,madani,pmx,poli
6
  ECONOMIC: ekonomi,economy,economic,bank,banking,ringgit,rm,usd,dollar,euro,yen,pound,pelaburan,investment,invest,kewangan,finance,financial,bisnes,business,perdagangan,trade,trading,eksport,export,import,gdp,gross domestic product,kadar faedah,interest rate,inflasi,inflation,deflasi,deflation,saham,stock,shares,equity,bond,sukuk,mata wang,currency,forex,foreign exchange,bank negara,bnm,central bank,miti,ministry of international trade,bursa malaysia,klse,stock exchange,felda,federal land development,petronas,petroleum nasional,genting,maybank,malayan banking,cimb,commerce international,public bank,rhb,rashid hussain,hong leong,ammbank,ambank,alliance bank,affin bank,bsn,bank simpanan,agro bank,bank pertanian,bank islam,bimb,bank muamalat,ocbc,uob,standard chartered,hsbc,citibank,deutsche bank,bilion,billion,juta,million,ribu,thousand,ratus,hundred,tender,kontrak,contract,projek,project,syarikat,company,sdn bhd,sendirian berhad,bhd,berhad,plc,public limited,ltd,limited,korporat,corporate,industri,industry,manufacturing,pengilangan,teknologi,technology,digital,fintech,financial technology,startup
7
 
8
  # Law Keywords
9
- LAW: mahkamah,court,hakim,judge,undang-undang,law,legal,polis,police,sprm,macc,malaysian anti-corruption,anti-corruption,rasuah,corruption,jenayah,crime,criminal,kes,case,pendakwa,prosecutor,peguam,lawyer,attorney,solicitor,barrister,tribunal,tangkap,arrest,dakwa,charge,tuduhan,allegation,hukuman,sentence,penjara,prison,jail,suspek,suspect,tertuduh,accused,saksi,witness,bukti,evidence,ipcmc,independent police,agc,attorney general,peguam negara,chief justice,ketua hakim,federal court,mahkamah persekutuan,court of appeal,mahkamah rayuan,high court,mahkamah tinggi,sessions court,mahkamah sesyen,magistrate court,mahkamah majistret,syariah court,mahkamah syariah,industrial court,mahkamah perusahaan,juvenile court,mahkamah juvana,scam,penipuan,fraud,dadah,drugs,narkotik,narcotic,rompakan,robbery,samun,snatch theft,bunuh,murder,rogol,rape,khalwat,zina,adultery,syariah,islamic law,hudud,fatwa,mufti,imam,ustaz,religious teacher,enforcement,penguatkuasaan,investigation,siasatan,forensic,forensik
10
 
11
  # Danger Keywords
12
  DANGER: banjir,flood,kemalangan,accident,kebakaran,fire,covid,coronavirus,pandemic,wabak,epidemic,virus,influenza,denggi,dengue,malaria,tuberculosis,tb,cancer,kanser,heart attack,serangan jantung,stroke,diabetes,kencing manis,hypertension,darah tinggi,gempa,earthquake,tsunami,bahaya,danger,dangerous,darurat,emergency,bencana,disaster,catastrophe,mangsa,victim,casualties,korban,maut,death,meninggal,die,cedera,injured,luka,wound,hospital,ambulans,ambulance,letupan,explosion,bomb,bom,teroris,terrorist,terrorism,keganasan,jpam,civil defence,bomba,fire department,rescue,menyelamat,evakuasi,evacuate,shelter,tempat perlindungan,landslide,tanah runtuh,haze,jerebu,pollution,pencemaran,toxic,toksik,chemical,kimia,radiation,radiasi,nuclear,nuklear,radioactive,radioaktif,leak,bocor,spill,tumpahan,contamination,pencemaran,poisoning,keracunan
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
  ECONOMIC: ekonomi,economy,economic,bank,banking,ringgit,rm,usd,dollar,euro,yen,pound,pelaburan,investment,invest,kewangan,finance,financial,bisnes,business,perdagangan,trade,trading,eksport,export,import,gdp,gross domestic product,kadar faedah,interest rate,inflasi,inflation,deflasi,deflation,saham,stock,shares,equity,bond,sukuk,mata wang,currency,forex,foreign exchange,bank negara,bnm,central bank,miti,ministry of international trade,bursa malaysia,klse,stock exchange,felda,federal land development,petronas,petroleum nasional,genting,maybank,malayan banking,cimb,commerce international,public bank,rhb,rashid hussain,hong leong,ammbank,ambank,alliance bank,affin bank,bsn,bank simpanan,agro bank,bank pertanian,bank islam,bimb,bank muamalat,ocbc,uob,standard chartered,hsbc,citibank,deutsche bank,bilion,billion,juta,million,ribu,thousand,ratus,hundred,tender,kontrak,contract,projek,project,syarikat,company,sdn bhd,sendirian berhad,bhd,berhad,plc,public limited,ltd,limited,korporat,corporate,industri,industry,manufacturing,pengilangan,teknologi,technology,digital,fintech,financial technology,startup
7
 
8
  # Law Keywords
9
+ LAW: mahkamah,court,hakim,judge,undang-undang,law,legal,polis,police,sprm,macc,malaysian anti-corruption,anti-corruption,rasuah,corruption,jenayah,crime,criminal,kes,case,pendakwa,prosecutor,peguam,lawyer,attorney,solicitor,barrister,tribunal,tangkap,arrest,dakwa,charge,tuduhan,allegation,hukuman,sentence,penjara,prison,jail,suspek,suspect,tertuduh,accused,saksi,witness,bukti,evidence,ipcmc,independent police,agc,attorney general,peguam negara,chief justice,ketua hakim,federal court,mahkamah persekutuan,court of appeal,mahkamah rayuan,high court,mahkamah tinggi,sessions court,mahkamah sesyen,magistrate court,mahkamah majistret,syariah court,mahkamah syariah,industrial court,mahkamah perusahaan,juvenile court,mahkamah juvana,scam,penipuan,fraud,dadah,drugs,narkotik,narcotic,rompakan,robbery,samun,snatch theft,bunuh,murder,rogol,rape,khalwat,zina,adultery,syariah,islamic law,hudud,fatwa,mufti,imam,ustaz,religious teacher,enforcement,penguatkuasaan,investigation,siasatan,forensic,forensik,pemandu,driver,lalu lintas,traffic,jalan raya,road,highway,kenderaan,vehicle,kereta,car,motosikal,motorcycle,bas,bus,lori,truck,van,jpj,road transport,jabatan pengangkutan jalan,sumaman,summons,kompaun,compound,lesen,license,permit,undang-undang jalan raya,road traffic act,akta pengangkutan jalan,road transport act,telefon bimbit,mobile phone,handphone,hp,telefon,phone,lampu isyarat,traffic light,isyarat,signal,trafik,parking,letak kereta,had laju,speed limit,laju,speed,overtake,memotong,u-turn,pusing u,roundabout,bulatan,helmet,topi keledar,seat belt,tali pinggang keledar,drunk driving,memandu mabuk,alcohol,alkohol,breathalyzer,ujian nafas,roadblock,sekatan jalan,mata,police roadblock,tnb,tenaga nasional,utility,kemudahan awam,public utility,infrastruktur,infrastructure,jambatan,bridge,terowong,tunnel,lebuhraya,expressway,plus,projek lebuhraya,highway project
10
 
11
  # Danger Keywords
12
  DANGER: banjir,flood,kemalangan,accident,kebakaran,fire,covid,coronavirus,pandemic,wabak,epidemic,virus,influenza,denggi,dengue,malaria,tuberculosis,tb,cancer,kanser,heart attack,serangan jantung,stroke,diabetes,kencing manis,hypertension,darah tinggi,gempa,earthquake,tsunami,bahaya,danger,dangerous,darurat,emergency,bencana,disaster,catastrophe,mangsa,victim,casualties,korban,maut,death,meninggal,die,cedera,injured,luka,wound,hospital,ambulans,ambulance,letupan,explosion,bomb,bom,teroris,terrorist,terrorism,keganasan,jpam,civil defence,bomba,fire department,rescue,menyelamat,evakuasi,evacuate,shelter,tempat perlindungan,landslide,tanah runtuh,haze,jerebu,pollution,pencemaran,toxic,toksik,chemical,kimia,radiation,radiasi,nuclear,nuklear,radioactive,radioaktif,leak,bocor,spill,tumpahan,contamination,pencemaran,poisoning,keracunan
13
+ ```
14
+
15
+ ---
16
+
17
+ ## 🌐 **STEP 2: GO TO HUGGINGFACE**
18
+
19
+ 1. **Open:** https://huggingface.co/rmtariq/malaysian-priority-classifier
20
+ 2. **Click:** "Files and versions" tab
21
+ 3. **Click:** on `classification_rules.txt` file
22
+ 4. **Click:** the "Edit" button (pencil icon)
23
+
24
+ ---
25
+
26
+ ## ✏️ **STEP 3: REPLACE THE CONTENT**
27
+
28
+ 1. **Select All** the old content (Cmd+A)
29
+ 2. **Delete** it
30
+ 3. **Paste** the new content you copied above (Cmd+V)
31
+
32
+ ---
33
+
34
+ ## 💾 **STEP 4: SAVE THE CHANGES**
35
+
36
+ 1. **Commit message:** `Update v1.1: Enhanced traffic law keywords`
37
+ 2. **Click:** "Commit changes to main"
38
+
39
+ ---
40
+
41
+ ## 🧪 **STEP 5: TEST THE UPDATE**
42
+
43
+ **Test with this text:**
44
+ ```
45
+ Pemandu tidak boleh pegang dan guna telefon bimbit ketika berhenti di lampu isyarat
46
+ ```
47
+
48
+ **Expected result:** `Law` ✅
49
+
50
+ ---
51
+
52
+ ## 🎯 **WHAT'S NEW IN THE LAW SECTION:**
53
+
54
+ Look for these new keywords in the LAW line:
55
+ - **pemandu,driver** - Driver content
56
+ - **telefon bimbit,mobile phone** - Phone usage
57
+ - **lampu isyarat,traffic light** - Traffic signals
58
+ - **jpj,road transport** - Transport authority
59
+ - **sumaman,kompaun** - Traffic violations
60
+ - **lesen,license** - Driving permits
61
+
62
+ ---
63
+
64
+ ## ✅ **SUCCESS!**
65
+
66
+ After this update, your HuggingFace model will correctly classify traffic law content as **LAW** category, just like your local version!
67
+
68
+ **🚀 Your model is now enhanced with 50+ traffic law keywords!**