Eliciting Positive Emotional Impact in Dialogue Response Selection

摘要

Introduction of emotion into human-computer interaction (HCI) have allowed various system's abilities that can benefit the user. Among many is emotion elicitation, which is highly potential in providing emotional support. To date, works on emotion elicitation have only focused on the intention of elicitation itself, e.g. through emotion targets or personalities. In this paper, we aim to extend the existing studies by utilizing examples of human appraisal in spoken dialogue to elicit a positive emotional impact in an interaction. We augment the widely used example-based approach with emotional constraints: (1) emotion similarity between user query and examples, and (2) potential emotional impact of the candidate responses. Text-based human subjective evaluation with crowdsourcing shows that the proposed dialogue system elicits an overall more positive emotional impact, and yields higher coherence as well as emotional connection.