ID
int64
815,347,703B
1,542,658,424B
label
stringlengths
1
89
1,509,440,958,391,361,500
įˆ˜įˆįŠ«įˆ_į‰€įŠ•
1,445,881,508,682,735,600
GleanEthiopian
1,430,885,386,713,112,600
Gambella
1,472,174,049,052,741,600
į‹µįˆ
1,346,926,704,925,794,300
TigrayGenocide
1,347,994,497,502,871,600
įˆ°į‰ įˆ­_į‹œįŠ“
1,499,985,982,149,562,400
RejectHR6600
816,031,277,749,010,400
SouthSudan
1,183,270,021,910,413,300
FakeNewsAlert
1,528,211,634,862,231,600
ColorsOfUnity
1,472,261,557,811,880,000
FreeJawarMohammed
1,330,102,233,510,404,000
BisratTv
1,387,521,631,115,952,000
share
1,399,130,192,925,319,200
įŠ įˆœįˆŖįŠ«
1,528,046,001,487,073,300
AmharaUnderAttack
1,461,259,684,636,201,000
PanAfricanismPrevails
1,243,297,587,513,491,500
Turkey
1,471,536,859,801,133,000
į‹įŠ–
1,464,968,252,568,780,800
EritreaPrevails
1,506,300,293,188,923,400
AfarGenocide
1,535,911,607,653,609,500
ItsMyDam
1,459,195,894,520,332,300
įŠ¢į‰µį‹®įŒµį‹«
1,476,774,186,957,815,800
nomore
1,437,470,814,291,611,600
TPLFTerroristGroup
950,823,915,105,943,600
AU
1,421,055,483,599,892,500
įŠ įˆ»įˆ_į‹œįŠ“
1,490,087,756,567,044,000
SaveTigray
953,717,247,205,396,500
StarveTheTPLFBeast
1,270,218,230,779,588,600
BringBackOurGirls
1,443,834,403,231,342,600
įŠ¢įˆ¬į‰»
1,346,252,753,975,337,000
abiymustlead
1,532,392,965,846,487,000
NOMORE
1,475,492,315,959,644,200
Share
1,542,259,292,040,052,700
įŠ„įŠ“įˆ˜įˆ°įŒįŠ“įˆˆįŠ•
1,456,167,098,871,423,000
ShegerWerewoch
1,491,787,868,120,912,000
į‹œįŠ“
1,504,503,975,185,424,400
TikTok
1,512,528,034,233,991,200
EACC
1,374,858,951,209,128,000
į‰ įŠ¢į‰µį‹®įŒµį‹«
1,520,458,350,818,902,000
GondarMassacre
1,438,162,495,890,460,700
įˆ°į‰ įˆ­_į‹œįŠ“
1,444,970,161,153,380,400
įŠ¢į‰µį‹®įŒµį‹«_įŠ įˆøįŠ•į‹įˆˆį‰½
1,538,954,404,195,405,800
įŠ įˆØįŠ•įŒ“į‹“_įŠ įˆ»įˆ«
1,349,776,945,245,020,200
StopWarOnTigray
1,305,588,205,069,185,000
OLF
1,181,113,503,253,483,500
EPRDF
1,516,870,426,814,730,200
BreakTheSilence
1,056,710,920,129,384,400
įŠ į‰„į‹­
1,392,006,717,542,740,000
WBO
1,521,196,162,765,664,300
AmharaUnity
1,406,673,827,048,968,200
ForwardElectionEthiopia
1,276,827,747,281,703,000
įŠ įįˆŖįŠ«
1,467,551,581,369,299,000
no
1,515,444,247,759,052,800
OLA
1,431,017,606,027,305,000
UnityForEthiopia
1,459,957,990,090,588,200
TplfIsWarCriminal
1,060,081,500,446,699,500
įŠ¦į‹«į‹«
1,391,620,779,750,592,500
EthiopiaCheck
1,488,949,733,968,257,000
covid19ethiopia
1,384,374,128,707,256,300
Thedaysaretooshort
1,256,820,419,803,349,000
EthioįŒµia
1,326,952,294,617,845,800
TplfmustGo
1,380,116,675,429,818,400
Committed_to_strengthening_the_lasting_unity_of_the_people
1,502,237,811,361,124,400
StopBombingTigray
1,407,633,636,539,654,100
įŠ į‰¦įˆ
1,454,836,222,132,641,800
UnityForEthio
1,047,441,387,971,653,600
EPRDF
1,296,229,849,573,412,900
OromoProtests
1,258,982,862,096,908,300
įˆįˆ­įŒ«
1,442,574,169,187,704,800
Ethiopia
1,419,712,773,924,212,700
TigrayIsPrevailing
1,538,930,524,085,014,500
įŠ įˆØįŠ•įŒ“į‹“įŠ įˆ»įˆ«
1,494,674,245,045,170,200
į‰„įˆįŒ½įŒįŠ“
1,418,561,196,031,103,000
į‰µįŒįˆ«į‹­
1,433,478,027,254,669,300
NoNegotiationWithTplf
1,491,346,280,320,278,500
StopTigraySiege
1,465,709,028,290,404,400
įŠ¢į‰µį‹®įŒµį‹«į‹Š
1,432,624,711,175,975,000
į‰ įŠ įˆ›įˆ«
1,522,532,740,062,756,900
AmharaGenocide
1,456,155,155,313,930,200
įŠ į‹²įˆµįŠ į‰ į‰£
1,520,100,516,927,357,000
Prosperity
1,474,752,593,213,726,700
į‹ØįŠ¢į‰µį‹®įŒµį‹«įŠ•
1,406,671,353,567,232,000
UnityForEthiopia
1,416,059,879,341,580,300
TigrayMassArrest
1,436,705,959,619,940,400
į‰µįŒįˆ«į‹­į‰µįˆµį‹•įˆ­
1,471,798,099,622,436,900
Tigray
1,465,425,624,827,318,300
LearnifyEdTech
1,456,260,143,578,898,400
addisababa
1,431,948,415,466,885,000
EthiopiaPrevails
1,466,790,443,232,936,000
TplfIsWarCriminal
1,481,388,964,649,054,200
į‹ˆį‹«įŠ”
1,385,202,135,013,503,000
įŠ¦į‹«į‹«įˆ˜įˆį‰²įˆšį‹²į‹«
1,247,948,550,408,933,400
COVID19Ethiopia
1,511,746,011,089,318,000
RejectHR6600
1,512,242,800,481,419,300
SupportHR6600
821,438,982,609,367,000
EthiopianRevolution
1,384,372,438,327,234,600
AmharaGenocide
1,396,878,938,190,880,800
EritreaShinesAt30
901,090,343,977,975,800
VOAAmharic
1,276,151,729,013,358,600
itsmydam

Hashtag Prediction Dataset from paper TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations

PRs Welcome arXiv Github

This repo contains the Hashtag prediction dataset from our paper TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations.
[arXiv] [HuggingFace Models] [Github repo]

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Download

Use the hashtag-classification-id.zip in this repo. Link.

Check the first-author's GitHub repo for any supplemental dataset material or code. Link

Dataset Description

The hashtag prediction dataset is a multilingual classification dataset. Separate datasets are given for different languages. We first select 500 (or all available) popular hashtags of each language and then sample 10k (or all available) popular Tweets that contain these hashtags. We make sure each Tweet will have exactly one of the selected hashtags.

The evaluation task is a multiclass classification task, with hashtags as labels. We remove the hashtag from the Tweet, and let the model predict the removed hashtag.

We provide Tweet ID and raw text hashtag labels in tsv files. For each language, we provide train, development, and test splits.

To use the dataset, you must hydrate the Tweet text with Twitter API, and remove the hashtag used for label from each Tweet .

The data format is displayed below.

ID label
1 hashtag
2 another hashtag

Citation

If you use our dataset in your work, please cite the following:

@article{zhang2022twhin,
  title={TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations},
  author={Zhang, Xinyang and Malkov, Yury and Florez, Omar and Park, Serim and McWilliams, Brian and Han, Jiawei and El-Kishky, Ahmed},
  journal={arXiv preprint arXiv:2209.07562},
  year={2022}
}
Downloads last month
52