tweet
stringlengths
6
380
label
class label
2 classes
bhaunk bhaunk ke pagal ho jaata ab tak....
0Abusive/Offensive
shivsaina nam mulla saina rakh hijra uddhav thackeray
0Abusive/Offensive
there are some yahodi daboo like imran chore zakat khore
0Abusive/Offensive
chutiye ye hunduism ni he islam he , isme koi kisi muslim ko ni maarta gashti ke bache
0Abusive/Offensive
han g rakhail e awal kiya bak rai hai. kisay ahmaq kah rai hai dali. tera baap ho ga ahmaq ghasti bosri ki dali
0Abusive/Offensive
can you please stop pinning a fan everytime bro what you need attention ki log aakar gaali de usne tera naam likha ho post me usne tere idol sid ko kuch bola ho ajeeb chutiya ho yaar poke krna h 24ghante tum chutiyo ko show ko show ki tarah dekho bc
0Abusive/Offensive
hum kisi se km nhi
1Normal
aurat bhatak sakri hy kbi β€œmaa” ni bhatakti . ye line hmy btati hy ky islam ny nikah ko kitna khubsurat rishta bnaya hy , jismy aurat phly sy ziada mazboot hojati , mukammal hojati hy . β™₯️ #alif
1Normal
ye ek aur a gya...bikkau maal...mufty kay kutay...mujib shami lifafay k sath beth kr putar tum jo ho wo chupp nahi sakta... university mein tum jesy londday mushkil waqt mein kaam aatay hain...useless stunts
0Abusive/Offensive
akhrott wants to know your location..
0Abusive/Offensive
with due respect sir kya aurat march se ye chezain khtm hu jain gi...?
1Normal
.. mera jism meri marzi ka nara woh aurteen laga rahi hain jinki marzi ho betu mard hath bandh kar mazrat kar leta hai.πŸ™πŸ»πŸ˜”#stopauratbarbadimarch ..
0Abusive/Offensive
good next time b kabhi koi chizz mat mangnnna..... coz we have mind not heart.... stay blessed 😘
0Abusive/Offensive
hahaha ap ke jawab bta rahe kon gussa kr raha ha aur haqiqat ko jutlaya nahi ja sakta api boot chatney wali riwayat n leauge me paie jati ha 😊😊😊😊
0Abusive/Offensive
#boycottchhapaak deepika randi hai
0Abusive/Offensive
πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚πŸ˜‚ chal ja tujhy maaf kia
1Normal
agr ap chahtay hn k log ap ko acha kahean to apna naam acha rakh lean or koi tareeqa nai h
1Normal
nhi walid sabh ne wo aksr tumri maa sey milne aty te.
0Abusive/Offensive
ooper se sir g msla ye ha k this works too πŸ˜πŸ˜‚
1Normal
apk options thk ni hain....basically logon ko march pe ni slogans pe aitraz hai
1Normal
... want or change their religion. #pakistan’s constitution does not give its citizens right to choose any religion or even change it, in fact it criminalizes and persecutes its own ahmaddiya muslim community (who are criminalized to not even call themselves muslims)
1Normal
abay saalay
0Abusive/Offensive
haaan mene b dekha thaa ye matermind nai dhokebaz hy
0Abusive/Offensive
no we do not want a syria. peoples sects should he stated instead of accusations being fired hes ahmedi. etc . we are a sunni country it should remain that way.
1Normal
aray bosdk 🌚 ✌️
0Abusive/Offensive
abi tk tu khair hi haaπŸ˜‚
1Normal
hn ye sai hy πŸ˜‚πŸ˜‚πŸ˜‚
1Normal
teri baji ko bhe ye he aitraz ha ha wesay patwari khota he tera asli abba ha beshak apni ami jan say puch lay baki imran yahodi bayghart kashmir frosh say to teri ami jan nay taste change kia ha imrani tati sahib
0Abusive/Offensive
masjid nhi tm mullo ka kabr khodunga wha madarchod halale ki paidaish ...10 baap ki aulaad .. pakistani kutte saale .. hindustan m rhte ho hm hindu log tmhare maa bhen chod denge bhen chod agr hmari dimag ki hti to
0Abusive/Offensive
pechy auntie check kar bhai iskaa link milyy ga kia πŸ˜œπŸ˜œπŸ‘†
0Abusive/Offensive
haan chal abb teri baari😁
1Normal
me right now: ye sun'ny k liye intzar kia tha! πŸ˜‘πŸ˜­ #pslv #psl2020 #tyyarhain
1Normal
tum jesi ghatia aurat ka kam hi ni hai media par.kisi kothhay pe ja aur ulta late ja ke.begherat
0Abusive/Offensive
bhai, koi tip hi de dyn phr aap? πŸ˜‚πŸ˜‚
1Normal
haha randi saali geeli hoae tu
0Abusive/Offensive
rt : tf is wrong with arsenal. why the helll are you building from the backkkkk bhenchod where are your braincellss
0Abusive/Offensive
mickey mouse dakho chup kar k 😀
0Abusive/Offensive
rt : bhenchod pakad pakad ke musalmano se bayan le rahe hain πŸ˜¬πŸ˜‚πŸ˜‚
0Abusive/Offensive
bat bhagnaay ki nahi hai randi tu hai randi tujhsy to achay achay side pe hotay hai magar koi koi teri maa chodta hai un mai sy aj mai ek hogya jo teri maa chod gya
0Abusive/Offensive
gandu dimag bahanchod senseless baatein bosdkbosdk likhne se phle soch to
0Abusive/Offensive
you need to research how many lands islam invaded in the 200 years after the hijra.
1Normal
aurat ho gy tum 😑😑
0Abusive/Offensive
hello aunty he is just hitting his own thigh kon si aankh se dekh rahi yaa jyaad ho gya eeeks nakal gya rumhara saali c grade bitch πŸ˜‘πŸ˜‘πŸ–•πŸ»πŸ–•πŸ»πŸ–•πŸ»
0Abusive/Offensive
ja bibi maaf kar..🀣🀣
1Normal
dekh logo ko cheat b krre h ... or ek bhenchod hmara to koi katne ko b tyaar nhi πŸ₯Ί
0Abusive/Offensive
jab hogi tab bhonkna yaha pe kukku name pta karlena punjab matlab btau sidhe shabdo me ( rand )
0Abusive/Offensive
hahahah rameen api pata nahi kahin parha tha id remember. khet ap keh rhi hyn thik hi hoga πŸ™ŠπŸ˜‚
1Normal
buht zaberdast choice ha apke πŸ‘Œ
1Normal
ji o chodry ustaad lgga en... india di pehn di siri
0Abusive/Offensive
nai nai nai bilkul bhi nai matlab tum ho kon jiski my ny tension leni hy wo to aesy hi my msgs kar rahi thi ky meryum sahiba wapes a jain
1Normal
lanti insan google kro phr bakwas kro ahlebait jan qurban hy tumhari trha nhe bughz main bhray hoay
0Abusive/Offensive
admission ke liye tumhare pass i'd hai ,free ka gas ,home and so many nrc me tumhe kya problem yadi itna hi problems hain to pakistan bhag bhodsri saali
0Abusive/Offensive
rt : black friday is yahoodi name.. so we are celebrating white/big friday 😝format yahoodi ho name yahoodi nahe hona chaye πŸ˜‚
1Normal
😏😏😏 noi g mai backbencher thi
1Normal
harrami police rishwat khore like imran chore zakat khore yahodi agents both are bajwa and imran chore destroyed pakistan and business of pakistan nation. lakh lanat teray jaman tay harrami qadiyani
0Abusive/Offensive
tukde tukde gang ke baare mein bhi bhaunk le kuch kutte...apne baap kanhaiya chutiye ke baare mein bol...keju kutte ke baare mein bol...namo forever πŸ‘πŸ‘
0Abusive/Offensive
yeh bh...ri ka mch bahut bhaunk raha hai now needs castration
0Abusive/Offensive
soch rha hn danish bhuk ki waja sy mara ha?
1Normal
phir tau aap dab sae bari ahmaq hain kay aapke mutabaq jo chief ahmaq hai uss say tau baray shauq say shadi kar lo thi.
0Abusive/Offensive
trri pen ki chooli q jalti ha randi k bachy tera kia kaam army k mamlaat mein bolny ka , or haram k pillay tum lpg deshat ghard ho harami afghani nikalo humary mulk say , tum ne apny haramiyun k sath firing kr k q jawan shaheed kiay ? kis yahodi k tookron pe palta ha tu kuttay
0Abusive/Offensive
cigrette vale pe b blush kiye thee
1Normal
ps: main bhi small handle hoon.πŸ™πŸ™ abhi abhi ek bhai saheb taana maar ke gaye hain πŸ˜€πŸ˜€
1Normal
hmain tou nzr nai ain kbhi teri storiesπŸ˜œπŸ˜‚πŸ˜‚
1Normal
sardar mat bol isko yr or kuch bhi bol ganda sa sardar nhi hai aurat ki respect nhi karta asli nhi hai
0Abusive/Offensive
aap 11 se pehle daal lia krien us me jo kch dalna ho
1Normal
wah g munafiqat pm to charsi he hai hamara jaha marzi ho banda pm hamara woi rahey ga apeal be usi ko kerni
0Abusive/Offensive
atankwadi pragya kya wo tere maa baap ko bomb se udai kya jo tu chutiya paida hua or digvijay singh jo rishte me tera baap lagta h usne bola tha hafiz sahid ji osama ji samjha n gutterchhap.
0Abusive/Offensive
date pe bill khud pay kro. #stopauratbarbadimarch
1Normal
pattharbaazi aur public property wahi damage karte jo 10-10 bachche paida karke 1 paise ka tax na dete. jinki jeb se paisa jaata unhe hi kadar hoti. abhi jo tax dete wo dikhayenge 2024 mein 2019 ka repeat telecast πŸ˜‚
1Normal
q kasa diya .#coronarvirus bhai please bachoo is virus se #help karo bus ki.#godsaveus
1Normal
hahaha pahly sochun ga phir nah kah dunga
1Normal
kis k jahaz pe gai sawal ye hai😜😜😜😜😜
1Normal
bhai aap bhi samjho. sirf 20 seconds mein sara drama pura. media coverage walon ko koi darr nhi. gajjab!!!! πŸ€”πŸ€”πŸ€”πŸ€”πŸ€”
1Normal
mujhe tm par fakhr h tumhe main bohat bohat pasnd karta ho i really like you agr meri baat achi lage to reply karna
1Normal
bahanchod kash us din teri maa ne oral hi kr lia hota
0Abusive/Offensive
lun pr charey apni marzi se aye hain bharwe
0Abusive/Offensive
le ab toh effect mein bhi aa gaya...ab ek kar burnol laga le...🀣🀣🀣🀣🀣🀣 kuch ukhad payi kya abhi tak pakistani suariya...tere saath ke kutte bhi bas bhaunk paye ...
0Abusive/Offensive
na andha na bahra pyar chalak hota ha 🎢
1Normal
rao jee apke peechay konsi mehwish parr gayi hai.
1Normal
allah ki mehrbani se dushman per lanat
0Abusive/Offensive
nicknames of my friends are quite unique. 1. basantiπŸ’ƒ 2. charsiπŸ‘Ί 3. billo😸 4. saddiqui chowk😜 5. neeli🀣
1Normal
dekh sab ko pata ke ab tu mujhe yaad nii krtiπŸ™‚
1Normal
g bilkul laanat ho aesi kartuton par..!! this is clear violation of military sops. budhhay ko sharam karni chahiye thori si..!
0Abusive/Offensive
tumse to wo bhi nhi bach pa rahi 🀣🀣🀣
1Normal
kon hai headr mein ? πŸ‘€
1Normal
wasy ek cheze nahi kahi pm ny pakistani o ghabrana nahi ha
1Normal
chandrachud ki sahi li jaa rahi hai, lekin ye bhenchod nazeer ko free pass kaise mil raha hai?
0Abusive/Offensive
apki behuda soch kisi b logo k pichy chupy iska pata logo ko chal hi jana hy, bs dukh is baat ka hy k hamary kuch bewaqoof log bina sochy aap jesi aurato ki bato me ajaty hyn.
0Abusive/Offensive
rt : are jo baag mein betthe hai unko samjho, tum ko samj mein aa gaya to kya laddu bante 😊😊😁😁
1Normal
na na koi buri bat nhi khudh bhi stalk krti aap
1Normal
ye kia bolrhy janab. pcb na hire ki ha through bid , tu they have gone through some bid at first. wo dainga na paisay pcb ko , aue khud wo ad's sa kamah ga aur viewershop etcetc sa
1Normal
aby gashti aurat apni mou say b allah ka naam na lay
0Abusive/Offensive
sc tou abhi bhi pro niazi hai , niazi ki khwahish par ns ko disqualify kya , wahan case kyon nahi jata niazi apna randi rona lay kar?
0Abusive/Offensive
yahan ki fizaaaa randi hai mitti to phir andaza laga lo
0Abusive/Offensive
bahanchod pant ye bsdk deserve karta hai international cricket me. #indvban
0Abusive/Offensive
pta nahi is bry me mujhy.. dekhny wlai ankh py dipnd hota hai.. wo wesa he dekhta hai yan nzren bdl k..
1Normal
agreed. but you have to take the point "controversy" just like you mentioned him "daku". proof to ap k pas b nai mgr zuban zadd e am bat hy to keh di ap ny b.. same like ritchie ny b mention kia k aisi baten suni gai hain. leave this now
1Normal
phr tu tmko tension lene ki zarorat nii
1Normal
so bilowal calling him beghairat,all those mullah brigades called him qadiyani, yahoodi talking abt hiswife, who has nothing to do with politics, is justified but he cant say anything?? what a logic.he can do and say whatever the fuck he want to. he is pm. u r nothing but a zero
1Normal
πŸ˜‚πŸ˜‚πŸ˜‚ yeh tu bilkul. mery bety ki trha baat kr rha hi
1Normal

Dataset Card for roman_urdu_hate_speech

Dataset Summary

The Roman Urdu Hate-Speech and Offensive Language Detection (RUHSOLD) dataset is a Roman Urdu dataset of tweets annotated by experts in the relevant language. The authors develop the gold-standard for two sub-tasks. First sub-task is based on binary labels of Hate-Offensive content and Normal content (i.e., inoffensive language). These labels are self-explanatory. The authors refer to this sub-task as coarse-grained classification. Second sub-task defines Hate-Offensive content with four labels at a granular level. These labels are the most relevant for the demographic of users who converse in RU and are defined in related literature. The authors refer to this sub-task as fine-grained classification. The objective behind creating two gold-standards is to enable the researchers to evaluate the hate speech detection approaches on both easier (coarse-grained) and challenging (fine-grained) scenarios.

Supported Tasks and Leaderboards

  • 'multi-class-classification', 'text-classification-other-binary classification': The dataset can be used for both multi class classification as well as for binary classification as it contains both coarse grained and fine grained labels.

Languages

The text of this dataset is Roman Urdu. The associated BCP-47 code is 'ur'.

Dataset Structure

Data Instances

The dataset consists of two parts divided as a set of two types, Coarse grained examples and Fine Grained examples. The difference is that in the coarse grained example the tweets are labelled as abusive or normal whereas in the fine grained version there are several classes of hate associated with a tweet.

For the Coarse grained segment of the dataset the label mapping is:- Task 1: Coarse-grained Classification Labels 0: Abusive/Offensive 1: Normal

Whereas for the Fine Grained segment of the dataset the label mapping is:- Task 2: Fine-grained Classification Labels 0: Abusive/Offensive 1: Normal 2: Religious Hate 3: Sexism 4: Profane/Untargeted

An example from Roman Urdu Hate Speech looks as follows:

{
  'tweet': 'there are some yahodi daboo like imran chore zakat khore'
  'label': 0
}

Data Fields

-tweet:a string denoting the tweet which has been selected by using a random sampling from a tweet base of 50000 tweets to select 10000 tweets and annotated for the dataset.

-label:An annotation manually labeled by three independent annotators, during the annotation process, all conflicts are resolved by a majority vote among three annotators.

Data Splits

The data of each of the segments, Coarse Grained and Fine Grained is further split into training, validation and test set. The data is split in train, test, and validation sets with 70,20,10 split ratio using stratification based on fine-grained labels.

The use of stratified sampling is deemed necessary to preserve the same labels ratio across all splits.

The Final split sizes are as follows:

Train Valid Test 7209 2003 801

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

The dataset was created by Hammad Rizwan, Muhammad Haroon Shakeel, Asim Karim during work done at Department of Computer Science, Lahore University of Management Sciences (LUMS), Lahore, Pakistan.

Licensing Information

The licensing status of the dataset hinges on the legal status of the Roman Urdu Hate Speech Dataset Repository which is under MIT License.

Citation Information

@inproceedings{rizwan2020hate,
  title={Hate-speech and offensive language detection in roman Urdu},
  author={Rizwan, Hammad and Shakeel, Muhammad Haroon and Karim, Asim},
  booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  pages={2512--2522},
  year={2020}
}

Contributions

Thanks to @bp-high, for adding this dataset.

Downloads last month
73
Edit dataset card

Models trained or fine-tuned on roman_urdu_hate_speech