Multilingual Twitter Corpus and Baselines for Evaluating Demographic Bias in Hate Speech Recognition

Existing research on fairness evaluation of document classification models mainly uses synthetic monolingual data without ground truth for author demographic attributes. In this work, we assemble and publish a multilingual Twitter corpus for the task of hate speech detection with inferred four author demographic factors: age, country, gender and race/ethnicity. The corpus covers five languages: English, Italian, Polish, Portuguese and Spanish. We evaluate the inferred demographic labels with a crowdsourcing platform, Figure Eight. To examine factors that can cause biases, we take an empirical analysis of demographic predictability on the English corpus. We measure the performance of four popular document classifiers and evaluate the fairness and bias of the baseline classifiers on the author-level demographic attributes.

[1]  Lucas Dixon,et al.  Ex Machina: Personal Attacks Seen at Scale , 2016, WWW.

[2]  Xuanjing Huang,et al.  How to Fine-Tune BERT for Text Classification? , 2019, CCL.

[3]  Sérgio Nunes,et al.  A Hierarchically-Labeled Portuguese Hate Speech Dataset , 2019, Proceedings of the Third Workshop on Abusive Language Online.

[4]  Gianluca Stringhini,et al.  Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior , 2018, ICWSM.

[5]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[6]  Tomas Mikolov,et al.  Enriching Word Vectors with Subword Information , 2016, TACL.

[7]  A G N,et al.  Bibliographical References , 1965 .

[8]  John C. Henderson,et al.  MITRE at SemEval-2019 Task 5: Transfer Learning for Multilingual Hate Speech Detection , 2019, SemEval@NAACL-HLT.

[9]  Saif Mohammad,et al.  Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems , 2018, *SEMEVAL.

[10]  Yoshua Bengio,et al.  Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling , 2014, ArXiv.

[11]  Daniel Jurafsky,et al.  Word embeddings quantify 100 years of gender and ethnic stereotypes , 2017, Proceedings of the National Academy of Sciences.

[12]  Carolyn Penstein Rosé,et al.  Perceptions of Censorship and Moderation Bias in Political Debate Forums , 2018, ICWSM.

[13]  Timnit Gebru,et al.  Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification , 2018, FAT.

[14]  Dirk Hovy,et al.  Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter , 2016, NAACL.

[15]  Adam Tauman Kalai,et al.  Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings , 2016, NIPS.

[16]  Jieyu Zhao,et al.  Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints , 2017, EMNLP.

[17]  Lucy Vasserman,et al.  Measuring and Mitigating Unintended Bias in Text Classification , 2018, AIES.

[18]  Gary King,et al.  Logistic Regression in Rare Events Data , 2001, Political Analysis.

[19]  Anne Marie Piper,et al.  Addressing Age-Related Bias in Sentiment Analysis , 2018, CHI.

[20]  Yejin Choi,et al.  The Risk of Racial Bias in Hate Speech Detection , 2019, ACL.

[21]  R'emi Louf,et al.  HuggingFace's Transformers: State-of-the-art Natural Language Processing , 2019, ArXiv.

[22]  Cristina Bosco,et al.  An Impossible Dialogue! Nominal Utterances and Populist Rhetoric in an Italian Twitter Corpus of Hate Speech against Immigrants , 2018, LREC.

[23]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[24]  David Yarowsky,et al.  Exploring Demographic Language Variations to Improve Multilingual Sentiment Analysis in Social Media , 2013, EMNLP.

[25]  Ingmar Weber,et al.  Racial Bias in Hate Speech and Abusive Language Detection Datasets , 2019, Proceedings of the Third Workshop on Abusive Language Online.

[26]  Alexandra Chouldechova,et al.  The Frontiers of Fairness in Machine Learning , 2018, ArXiv.

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

[28]  Udo Kruschwitz,et al.  Improving Hate Speech Detection with Deep Learning Ensembles , 2018, LREC.

[29]  Noah A. Smith,et al.  A Dependency Parser for Tweets , 2014, EMNLP.

[30]  Ankur Taly,et al.  Counterfactual Fairness in Text Classification through Robustness , 2018, AIES.

[31]  Lucy Vasserman,et al.  Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification , 2019, WWW.

[32]  F. Coulmas.  Sociolinguistics: The Study of Speakers' Choices , 2005 .

[33]  Petr Sojka,et al.  Software Framework for Topic Modelling with Large Corpora , 2010 .

[34]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[35]  Paolo Rosso,et al.  SemEval-2019 Task 5: Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter , 2019, *SEMEVAL.

[36]  Dirk Hovy,et al.  Cross-lingual syntactic variation over age and gender , 2015, CoNLL.

[37]  Pascale Fung,et al.  Reducing Gender Bias in Abusive Language Detection , 2018, EMNLP.

[38]  Soon-Gyo Jung,et al.  Assessing the Accuracy of Four Popular Face Recognition Tools for Inferring Gender, Age, and Race , 2018, ICWSM.

[39]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[40]  Yoon Kim,et al.  Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[41]  Wesley De Neve,et al.  Multimedia Lab @ ACL WNUT NER Shared Task: Named Entity Recognition for Twitter Microposts using Distributed Word Representations , 2015, NUT@IJCNLP.

[42]  Emily M. Bender,et al.  Data Statements for Natural Language Processing: Toward Mitigating System Bias and Enabling Better Science , 2018, TACL.

[43]  Mai ElSherief,et al.  Mitigating Gender Bias in Natural Language Processing: Literature Review , 2019, ACL.

[44]  Zeerak Waseem,et al.  Are You a Racist or Am I Seeing Things? Annotator Influence on Hate Speech Detection on Twitter , 2016, NLP+CSS@EMNLP.

[45]  Dirk Hovy,et al.  Demographic Factors Improve Classification Performance , 2015, ACL.

[46]  Carlos Gómez-Rodríguez,et al.  Language variety identification in Spanish tweets , 2014, EMNLP 2014.

[47]  Michael J. Paul,et al.  Neural User Factor Adaptation for Text Classification: Learning to Generalize Across Author Demographics , 2019, *SEM@NAACL-HLT.

[48]  Steven Bird,et al.  NLTK: The Natural Language Toolkit , 2002, ACL.

[49]  Svitlana Volkova,et al.  Inferring Latent User Properties from Texts Published in Social Media , 2015, AAAI.

[50]  Lyle H. Ungar,et al.  User-Level Race and Ethnicity Predictors from Twitter Text , 2018, COLING.

[51]  Thomas Hofmann,et al.  Leveraging Large Amounts of Weakly Supervised Data for Multi-Language Sentiment Classification , 2017, WWW.