Conversational Question Answering

Project Description

This project concerns conversational question answering (CQA), which is the task of answering a sequence of interrelated questions that occur during a conversation between a human and a computer. In contrast to answering unrelated single questions one question at a time, CQA is a more challenging task, due to the requirement of understanding a question in the context of previous questions and answers, as well as dealing with the effect of cascading errors when previous questions are incorrectly answered. We propose a novel approach using neural networks, which incorporates co-reference resolution to enable correct understanding of a subsequent question in the context of previous questions and answers.

Research Technical Area

  • Natural language processing (NLP)

Benefits to the society

The CQA algorithms developed in this project will provide the foundation to build the next generation of chatbots which can interact more naturally with humans. This will enable more widespread use of chatbots to serve the general public and customers, leading to improved productivity with far-reaching social and economic impact.


Project’s Publications

  1. Wee Chung Gan, Hwee Tou Ng.
    Improving the Robustness of Question Answering Systems to Question Paraphrasing. ACL 2019. 6065-6075. 10.18653/v1/P19-1610.

Team’s Principal Investigator

Professor Hwee Tou Ng
School of Computing
National University of Singapore

Professor Hwee Tou NG is Provost’s Chair Professor of Computer Science at the National University of Singapore (NUS) and a Senior Faculty Member at the NUS Graduate School for Integrative Sciences and Engineering. He received a Ph.D. in Computer Science from the University of Texas at Austin, USA. His research focuses on Natural language processing (NLP) and information retrieval. He is a Fellow of the Association for Computational Linguistics (ACL).

Recent Notable Awards

  • Fellow of the Association for Computational Linguistics (ACL), 2012.
  • Best Paper Award, 2011 Conference on Empirical Methods in Natural Language Processing (EMNLP 2011).