Proactive Intelligence in Conversational Assistants

Conversational assistants are getting more and more integrated with our everyday lives. However, the interaction with most of the current systems is one sided: the user asks a question and expects the system to respond. The goal of this project is to devise conversational systems that can proactively ask clarification questions when they are not certain about what the user is looking for. Two primary research questions we will investigate are: 1) when to ask the clarification question for the best user experience, and 2) what question to ask, and how to express this question. This candidate has already done some previous work related to this research area, has published in the area and has been attending various related workshops. Hence he is a very good fit to this research project.