Predictive Engagement: An Efficient Metric For Automatic Evaluation of Open-Domain Dialogue Systems

TL;DR

This paper introduces predictive engagement, a novel utterance-level metric for automatically evaluating open-domain dialogue systems, demonstrating its high correlation with human judgments and potential for real-time feedback.

Contribution

It proposes a new utterance-level engagement metric, predictive engagement, and shows its effectiveness in automatic evaluation and training of dialogue systems.

Findings

High agreement among human annotators on engagement scores

Conversation-level scores can be predicted from utterance-level scores

Utterance-level scores improve correlation with human judgments

Abstract

User engagement is a critical metric for evaluating the quality of open-domain dialogue systems. Prior work has focused on conversation-level engagement by using heuristically constructed features such as the number of turns and the total time of the conversation. In this paper, we investigate the possibility and efficacy of estimating utterance-level engagement and define a novel metric, {\em predictive engagement}, for automatic evaluation of open-domain dialogue systems. Our experiments demonstrate that (1) human annotators have high agreement on assessing utterance-level engagement scores; (2) conversation-level engagement scores can be predicted from properly aggregated utterance-level engagement scores. Furthermore, we show that the utterance-level engagement scores can be learned from data. These scores can improve automatic evaluation metrics for open-domain dialogue systems, as shown by correlation with human judgements. This suggests that predictive engagement can be used as a real-time feedback for training better dialogue models.