Research Interests: Social computing, natural language processing, machine learning, and communication.

At the CSG lab, the computational methods we develop augment the human knowledge for understanding benevolent online social movements and mitigating harmful online social content through the unique lens of social media. We curate socially impactful datasets, analyze them for a deepened understanding of behavior and linguistic patterns, and build Natural Language Processing (NLP) models that improve the classification and prediction of online misuse and community wellbeing.

Graduate Students (Current)

Undergraduate Students (Current)

ANNOUNCEMENT: I'm looking for students to work with at UCSD, including undergrad and grad students! If you're interested, please email me.

Publications [Google Scholar]

2021

• Latent Hatred: A Benchmark for Understanding Implicit Hate Speech.

M. ElSherief*, C. Ziems*, D. Muchlinski, V. Anupindi, J. Seybolt, M. De Choudhury, D. Yang. EMNLP 2021 [Dataset] (*Equal Contribution)

• Lifelong Learning of Hate Speech Classification on Social Media.

J. Qian, H. Wang, M. ElSherief, X. Yan. NAACL 2021

• 2020 and earlier

• Towards Understanding Gender Bias in Neural Relation Extraction.

A. Gaut, T. Sun, S. Tang, Y. Huang, J. Qian, M. ElSherief, J. Zhao, D. Mirza, E. Belding, K. Chang and W. Yang Wang . ACL 2020 [22%]

• Measuring and Characterizing Hate Speech on News Websites.

S. Zannettou, M. ElSherief, E. Belding, S. Nilizadeh, G. Stringhini (2020). WebSci 2020 [18%]

• Mitigating Gender Bias in Natural Language Processing: Literature Review.

T. Sun, A. Gaut, S. Tang, Y. Huang, M. ElSherief, J. Zhao, D. Mirza, E. Belding, Chang K., and W. Yang Wang. ACL 2019 [22%]

• Learning to Decipher Hate Symbols.

J. Qian, M. ElSherief, E. Belding, W. Yang Wang. NAACL 2019 [22%]

• J. Qian, M. ElSherief, E. Belding, W. Yang Wang (2018). Hierarchical CVAE for Fine- Grained Hate Speech Classification. In Proc. EMNLP 2018: Empirical Methods of Natural Language Processing [25%]

• J. Qian, M. ElSherief, E. Belding, W. Yang Wang (2018). Leveraging Intra-User and Inter-User Representation Learning for Automated Hate Speech Detection. In Proc. NAACL HLT 2018: North American Chapter of the Association for Computational Lin- guistics: Human Language Technologies [18%]

• M. ElSherief, V. Kulranki, D. Nguyen, W. Yang Wang, and E. Belding (2018). Hate Lingo: A Target-based Linguistic Analysis of Hate Speech in Social Media. In Proc. ICWSM 2018: AAAI Conference on Web and Social Media [16%]

• M. ElSherief, S. Nilizadeh, D. Nguyen, G. Vigna, and E. Belding (2018). Peer to Peer Hate: Hate Speech Instigators and Their Targets. In Proc. ICWSM 2018: AAAI Confer- ence on Web and Social Media [16%]

• M. ElSherief, B. Alipour, M. Al Qathrady, T. ElBatt, A. Zahran, and A. Helmy (2017). A Novel Mathematical Framework for Similarity-based Opportunistic Social Networks. In the Journal of Elsevier Pervasive and Mobile Computing.

• M. ElSherief, M. Vigil-Hayes, R. Raghavendra, and E. Belding (2017). Whom to Query? Spatially-Blind Participatory Crowdsensing under Budget Constraints. In ACM Workshop on Mobile Crowdsensing Systems and Applications, colocated with SenSys 2017.

• M. ElSherief, E. Belding, and D. Nguyen. (2017). #NotOkay: Understanding Gender- based Violence in Social Media. In Proc. ICWSM 2017: AAAI Conference on Web and Social Media. [14%]

• M. ElSherief, T. ElBatt, A. Zahran, and A. Helmy. (2015). An Information-theoretic Model for Knowledge Sharing in Opportunistic Social Networks. In Proc. SocialCom 2015: IEEE Conference on Social Computing and Networking. [25%]

• M. ElSherief and E. Belding. (2015). The Urban Characteristics of Street Harassment: A First Look. In ACM Workshop on Smart Cities and Urban Analytics, colocated with SigSpatial 2015.

• M. ElSherief, T. ElBatt, A. Zahran, and A. Helmy (2014). The Quest for User Similarity in Mobile Societies. In IEEE Workshop on Social and Community Intelligence, colocated with the Sixth IEEE/ASE International Conference on Social Computing (SocialCom) 2014. [17%]