Android dialogue system for customer service using prompt-based topic control and compliments generation

Abstract

This paper describes a dialogue system developed for the Dialogue Robot Competition 2023 that achieves topic control for trip planning by inserting text into prompts using the ChatGPT-API. We built a system that is capable of generating compliments for the user based on recognition of the user's appearance and creating travel plans by extracting the knowledge about the user's preference from the history of the user's utterances. Complements and planning based on preference are the elements required to maintain the quality of customer service. A preliminary round was held at a travel agency's actual store, where real customers experienced and evaluated the system. This system was evaluated first in the preliminary round and participated in the final round. The results of the preliminary round showed the effectiveness of the proposed system.