Ellery Wulczyn
ABSTRACT
The damage personal attacks cause to online discourse motivates many platforms to try to curb the phenomenon. However, understanding the prevalence and impact of personal attacks in online platforms at scale remains surprisingly difficult. The contribution of this paper is to develop and illustrate a method that combines crowdsourcing and machine learning to analyze personal attacks at scale. We show an evaluation method for a classifier in terms of the aggregated number of crowd-workers it can approximate. We apply our methodology to English Wikipedia, generating a corpus of over 100k high quality human-labeled comments and 63M machine-labeled ones from a classifier that is as good as the aggregate of 3 crowd-workers, as measured by the area under the ROC curve and Spearman correlation. Using this corpus of machine-labeled scores, our methodology allows us to explore some of the open questions about the nature of online personal attacks. This reveals that the majority of personal attacks on Wikipedia are not the result of a few malicious users, nor primarily the consequence of allowing anonymous contributions from unregistered users.
- J. Bergstra and Y. Bengio. Random search for hyper-parameter optimization. J. Mach. Learn. Res., 13:281--305, Feb. 2012. Google Scholar
Cross Ref
- M. Buhrmester, T. Kwang, and S. D. Gosling. Amazon's mechanical turk a new source of inexpensive, yet high-quality, data? Perspectives on psychological science, 6(1):3--5, 2011.Google Scholar
- J. Cheng, C. Danescu-Niculescu-Mizil, and J. Leskovec. Antisocial behavior in online discussion communities. In ICWSM, 2015.Google Scholar
- K. Dinakar, R. Reichart, and H. Lieberman. Modeling the detection of textual cyberbullying. The Social Mobile Web, 11:02, 2011.Google Scholar
- M. Duggan. Online harassment. Pew Research Center, 2014.Google Scholar
- Fundacion Barcelona Media (FBM). Caw 2.0 training datasets, 2009. http://caw2.barcelonamedia.org/.Google Scholar
- I. Gagliardone, D. Gal, T. Alves, and G. Martinez. Countering online hate speech. UNESCO Publishing, 2015.Google Scholar
- A. Halfaker. mwdiffs. https://github.com/mediawiki-utilities/python-mwdiffs.Google Scholar
- A. F. Hayes and K. Krippendorff. Answering the call for a standard reliability measure for coding data. Communication methods and measures, 1(1):77--89, 2007.Google Scholar
- Impermium. Detecting insults in social commentary dataset, 2012. https://www.kaggle.com/c/detecting-insults-in-social-commentary.Google Scholar
- K. Krippendorff. Content analysis: An introduction to its methodology. Sage, 2004.Google Scholar
- K. Krippendorff. Reliability in content analysis. Human communication research, 30(3):411--433, 2004.Google Scholar
- I. Kwok and Y. Wang. Locate the hate: Detecting tweets against blacks. In AAAI, 2013. Google ScholarDigital Library
- M. J. Moore, T. Nakano, A. Enomoto, and T. Suda. Anonymity and roles associated with aggressive posts in an online forum. Computers in Human Behavior, 28(3):861--867, 2012. Google ScholarDigital Library
- C. Nobata, J. Tetreault, A. Thomas, Y. Mehdad, and Y. Chang. Abusive language detection in online user content. In WWW, 2016. Google ScholarDigital Library
- B. Pang and L. Lee. Opinion mining and sentiment analysis. Foundations and trends in information retrieval, 2(1--2):1--135, 2008. Google ScholarDigital Library
- S. Pieschl, C. Kuhlmann, and T. Porsch. Beware of publicity! perceived distress of negative cyber incidents and implications for defining cyberbullying. Journal of School Violence, 14(1):111--132, 2015.Google ScholarCross Ref
- B. Plank, D. Hovy, and A. Søgaard. Learning part-of-speech taggers with inter-annotator agreement loss. In EACL, pages 742--751, 2014.Google ScholarCross Ref
- H. M. Saleem, K. P. Dillon, S. Benesch, and D. Ruths. A web of hate: Tackling hateful speech in online social spaces. In TA-COS, 2016.Google Scholar
- A. Schrock and D. Boyd. Problematic youth interaction online: Solicitation, harassment, and cyberbullying. Computer-Mediated Communication in Personal Relationships, pages 368--398, 2011.Google Scholar
- S. O. Sood, E. F. Churchill, and J. Antin. Automatic identification of personal insults on social news sites. Journal of the American Society for Information Science and Technology, 63(2):270--285, 2012. Google ScholarDigital Library
- N. Spirin and J. Han. Survey on web spam detection: principles and algorithms. ACM SIGKDD Explorations Newsletter, 13(2):50--64, 2012. Google ScholarDigital Library
- Support and Safety Team. Harassment Survey. Wikimedia Foundation, 2015. https://upload.wikimedia.org/wikipedia/commons/5/52/Harassment_Survey_2015_-_Results_Report.pdf.Google Scholar
- J. R. Tetreault, E. Filatova, and M. Chodorow. Rethinking grammatical error annotation and evaluation with the amazon mechanical turk. In NAACL-HLT, 2010. Google ScholarDigital Library
- R. S. Tokunaga. Following you home from school: A critical review and synthesis of research on cyberbullying victimization. Computers in human behavior, 26(3):277--287, 2010. Google ScholarDigital Library
- M. A. Walker, J. E. F. Tree, P. Anand, R. Abbott, and J. King. A corpus for research on deliberation and debate. In LREC, pages 812--817, 2012.Google Scholar
- W. Warner and J. Hirschberg. Detecting hate speech on the world wide web. In LSM, 2012. Google ScholarDigital Library
- Z. Waseem and D. Hovy. Hateful symbols or hateful people? predictive features for hate speech detection on twitter. In Proceedings of NAACL-HLT, pages 88--93, 2016.Google ScholarCross Ref
- D. Wiener. Negligent publication of statements posted on electronic bulletin boards: Is there any liability left after zeran. Santa Clara L. Rev., 39:905, 1998.Google Scholar
- Wikimedia. Harassment consultation 2015. https://meta.wikimedia.org/wiki/Harassment_consultation_2015.Google Scholar
- Wikimedia. Machine-learning tool to reduce toxic talk page interactions. https://meta.wikimedia.org/wiki/2015_Community_Wishlist_Survey/Bots_and_gadgets#Machine-learning_tool_to_reduce_toxic_talk_page_interactions.Google Scholar
- Wikipedia. Help:Talk pages. https://www.mediawiki.org/wiki/Help:Talk_pages.Google Scholar
- Wikipedia. Wikipedia:No personal attacks. https://en.wikipedia.org/wiki/Wikipedia:No_personal_attacks.Google Scholar
- Wikipedia. Wikipedia:Revision_deletion. https://en.wikipedia.org/wiki/Wikipedia:Revision_deletion.Google Scholar
- N. E. Willard. Cyberbullying and cyberthreats: Responding to the challenge of online social aggression, threats, and distress. Research Press, 2007. Google ScholarDigital Library
- E. Wulczyn, N. Thain, and L. Dixon. https://figshare.com/articles/Wikipedia_Detox_Data/4054689.Google Scholar
- G. Xiang, B. Fan, L. Wang, J. Hong, and C. Rose. Detecting offensive tweets via topical feature discovery over a large scale twitter corpus. In CIKM, 2012. Google ScholarDigital Library
- J.-M. Xu, B. Burchfiel, X. Zhu, and A. Bellmore. An examination of regret in bullying tweets. In HLT-NAACL, pages 697--702, 2013.Google Scholar
- M. L. Ybarra and K. J. Mitchell. Youth engaging in online harassment: Associations with caregiver--child relationships, internet use, and personal characteristics. Journal of adolescence, 27(3):319--336, 2004.Google ScholarCross Ref
- D. Yin, Z. Xue, L. Hong, B. D. Davison, A. Kontostathis, and L. Edwards. Detection of harassment on web 2.0. In WWW, 2009.Google Scholar
Index Terms
- Ex Machina: Personal Attacks Seen at Scale
Recommendations
Students' Sense of Belonging in Online Classes: Does Age Matter?
This study explored graduate students' perceptions of their sense of belonging in online classes, their comfort in participating in online discussions, and their technology skills. Differences in these areas among three different age groups were ...
Investigating the Motivational Paths of Peer Production Newcomers
CHI '17: Proceedings of the 2017 CHI Conference on Human Factors in Computing SystemsMaintaining participation beyond the initial period of engagement is critical for peer production systems. Theory suggests that an increase in motivation is expected with contributors' movement from the community periphery to the core. Less is known, ...
Comments