Muhammad Abdul-Mageed

I am an Assistant Professor at the iSchool, cross appointed with the Department of Linguistics. I am also an associate member of the Department of Computer Science. My areas of research interest are natural language processing, deep learning, Arabic natural language processing, and social media mining. My research is simply about teaching machines to understand and generate human language, with a focus on natural language socio-pragmatics (i.e., intelligent systems that understand and generate socially meaningful language). In my lab, our goal is to create intelligent ‘social’ machines that can interact naturally with humans. As an example application of my NLP work, we look at understanding and detecting bias and misinformation, specifically the context in which these happen. I am a founding member of UBC’s Center for Artificial Intelligence Decision making and Action (CAIDA). I teach courses in deep learning for natural language processing, python programming, sentiment analysis, and social media mining.

I have a dual Ph.D. in Computational Linguistics and Information Science from Indiana University, and was a Visiting Scholar in the Center for Computational Learning Systems at Columbia University from 2010 to 2012. I was also a Visiting Scholar in the University of Pennsylvania from 2016 to 2018.

I have received grants and funding from Google Research, The Natural Sciences and Engineering Research Council of Canada (NSERC), The Social Sciences and Humanities Research Council of Canada, The Fulbright Corporation, and UBC, and I have several software and patents for my work.

  • Natural language processing
  • Deep Learning
  • Natural language inference
  • Arabic
  • Social Artificial Intelligence
  • Misinformation
  • Social media mining

Kriegel, D., Abdul-Mageed, M., Robinson, D., Clark, J., Freeland, R., Heise, D., Rogers, K., and Smith-Lovin, L. (2017). A Multilevel Investigation of Arabic-Language Impression Change. International Journal of Sociology. 47(4).

Abdul-Mageed, M. (2017). Modeling Arabic Subjectivity and Sentiment in Lexical SpaceInformation Processing and Management.

 

Peer-reviewed papers

Elaraby, M., Abdul-Mageed, M.(2018). Deep Models for Arabic Dialect Identification on Benchmarked Data. In COLING VarDial.

Alhuzali, H., Abdul-Mageed, M., & Ungar, L (2018). Enabling Deep Learning of Emotion With First-Person Seed Expressions. In Proceedings of the Second Workshop on Computational Modeling of People’s Opinions, Personality, and Emotions in Social Media. NAACL HLT 2018, New Orleans, Louisiana, June 6, 2018.

Abdul-Mageed, M. (2018). Learning Subjective Language: Feature Engineered vs. Deep Models. In The 3rd Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT 2018), LREC2018.

Alshehri, A., Nagoudi, A., Alhuzali, H., and Abdul-Mageed, M. (2018). Think Before You Click: Data and Models for Adult Content in Arabic Twitter. In The 2nd Text Analytics for Cybersecurity and Online Safety (TA-COS-2018), LREC 2018, May 12, Miyazaki, Japan.

Abdul-Mageed, M., Alhuzali, H., and Elaraby, M. (2018). You Tweet What You Speak: A City-Level Dataset for Arabic Dialects. In LREC2018.

Abdul-Mageed, M., & Ungar, L. (2017). EmoNet: Fine-Grained Emotion Detection with Gated Recurrent Neural Networks. In ACL 2017.

Abdul-Mageed, M., Buffone, A., Peng, H., Giorgi, S., Eichstaedt, J., & Ungar, L. (2017). Recognizing Pathogenic Empathy in Social Media. In Proceedings of The 11th International AAAI Conference on Web and Social Media (ICWSM-17), May 15-18, Montreal, Canada.

 

 

Current Courses

Cross appointed with UBC Linguistics and Associate Member with Computer Science.