Discourse Analysis and Its Applications

Dr Shafiq Joty
Nanyang Technological University

Chaired by
Dr KAN Min Yen, Associate Professor, School of Computing

  20 Oct 2017 Friday, 03:00 PM to 04:00 PM

 Executive Classroom, COM2-04-02


A text (or conversation) is not merely a sequence of independent and isolated sentences, but instead a sequence of related sentences. It addresses a particular topic, often covering multiple subtopics, and is organized in a coherent way that enables the reader to process the information. In this talk, I will present computational models to uncover such linguistic structures of a given text (or conversation), and show their utility in downstream language processing applications. In particular, I will present structured models for predicting coherence structure of texts and dialogue structure of written asynchronous conversations (e.g., emails, forums), and show their applications in machine translation, text summarization and conversation thread recovery tasks.


Shafiq Joty is an assistant professor at the School of Computer Science and Engineering, NTU. Previously, he was a research scientist at the Qatar Computing Research Institute (QCRI). He holds a PhD in Computer Science from the University of British Columbia, where he was advised by Dr. Giuseppe Carenini, Dr. Raymond T. Ng, and Dr. Nando De Freitas. His work has primarily focused on developing discourse analysis tools (e.g., discourse parser, coherence model, topic model, dialogue act recognizer), and exploiting these tools effectively in downstream NLP applications like machine translation and summarization. Apart from discourse and its applications, he has also developed novel machine learning models for question answering, machine translation, and opinion analysis. He is also interested in interdisciplinary research, and has developed computational models for data mining tasks in social networks and in health science. His work has relied on deep learning for better representation of the input and on probabilistic graphical models for capturing dependencies in the output. His work has appeared in major journals and conferences such as CL, JAIR, TACL, CSL, ACL, EMNLP, NAACL, COLING, IJCAI, IUI, and ICWSM. Shafiq is a recipient of NSERC CGS-D scholarship and Microsoft Research Excellent Intern award.