Profiling Hate Speech Spreaders by Classifying Micro Texts Using BERT Model

Hate speech detection has lately gained considerable attention from researchers. To profile authors effectively, we consider classifying all tweets for a specific author independently. We used BERT, the pretrained model, to classify all individual tweets for each user. Then, we added an extra layer, called a confidence layer, by which we calculate the percentage of classified hateful tweets by the model and decide whether this author is spreading hate speech or not. We found this approach simple, yet effective in determining those considered haters. Our approach achieved 77% accuracy for the Spanish test dataset and 63% accuracy for the English test dataset.

[1]  Paolo Rosso,et al.  Overview of the Task on Automatic Misogyny Identification at IberEval 2018 , 2018, IberEval@SEPLN.

[2]  Paolo Rosso,et al.  Overview of PAN 2021: Authorship Verification, Profiling Hate Speech Spreaders on Twitter, and Style Change Detection - Extended Abstract , 2021, ECIR.

[3]  Viviana Patti,et al.  Resources and benchmark corpora for hate speech detection: a systematic review , 2020, Language Resources and Evaluation.

[4]  Felice Dell'Orletta,et al.  Overview of the EVALITA 2018 Hate Speech Detection Task , 2018, EVALITA@CLiC-it.

[5]  Lei Gao,et al.  Detecting Online Hate Speech Using Context Aware Models , 2017, RANLP.

[6]  Giovanni Vigna,et al.  Peer to Peer Hate: Hate Speech Instigators and Their Targets , 2018, ICWSM.

[7]  Paolo Rosso,et al.  Profiling Hate Speech Spreaders on Twitter Task at PAN 2021 , 2021, CLEF.

[8]  Benno Stein,et al.  TIRA Integrated Research Architecture , 2019, Information Retrieval Evaluation in a Changing World.

[9]  Viviana Patti,et al.  A New Measure of Polarization in the Annotation of Hate Speech , 2019, AI*IA.

[10]  Hugo Jair Escalante,et al.  Overview of MEX-A3T at IberLEF 2019: Authorship and Aggressiveness Analysis in Mexican Spanish Tweets , 2018, IberLEF@SEPLN.

[11]  Shervin Malmasi,et al.  Evaluating Aggression Identification in Social Media , 2020, TRAC.

[12]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[13]  Marco Guerini,et al.  CONAN - COunter NArratives through Nichesourcing: a Multilingual Dataset of Responses to Fight Online Hate Speech , 2019, ACL.

[14]  Shivakant Mishra,et al.  International Conference on Advances in Social Networks Analysis and Mining ( ASONAM ) Are They Our Brothers ? Analysis and Detection of Religious Hate Speech in the Arabic Twittersphere , 2018 .