Example Based Dialogue System Based on Satisfaction Prediction

Abstract

SummaryIn dialogue systems, dialogue modeling is one of the most important factors contributing to user satisfaction. Especially in example-based dialogue modeling (EBDM), effective methods for dialog example databases and selecting response utterances from examples improve dialogue quality. Conventional EBDM-based systems use example database consisting of pair of user query and system response. However, the best responses for the same user query are different depending on the user's preference. We propose an EBDM framework that predicts user satisfaction to select the best system response for the user from multiple response candidates. We define two methods for user satisfaction prediction; prediction using user query and system response pairs, and prediction using user feedback for the system response. Prediction using query/response pairs allows for evaluation of examples themselves, while prediction using user feedback can be used to adapt the system responses to user feedback. We also propose two response selection methods for example-based dialog, one static and one user adaptive, based on these satisfaction prediction methods. Experimental results showed that the proposed methods can estimate user satisfaction and adapt to user preference, improving user satisfaction score.

Exploratory Analysis of Usage Statistics of Dialogue Systems by Visualization

“…Studies on dialogue systems [5] in the fields of natural language processing and artificial intelligence have made significant progress. Many of these studies have focused on user support, including recommending information [6] , predicting user satisfaction [7], [8] , supporting decision-making [9] , education support [10], [11] , and providing health advisory systems [12] . These studies have contributed to solving user issues and increasing user satisfaction through dialogue systems.…”