Tour Concierge: Tour Route Recommendation System Based on User Short-Term and Long-Term Preference Models

Abstract

This chapter introduces Tour Concierge, an interactive system that learns each user's preferences and suggests optimal tour plans. A user preference model that is continuously optimized over time is proposed to recommend personalized tour routes. This model distinguishes between the user's short-term and long-term preferences and combines them with a tour route generation algorithm to increase user satisfaction. Twenty-six university students were asked to evaluate the use of this system and were surveyed. The survey results showed that 70.5% of the respondents felt that the tour routes generated by the system were compatible with their preferences. In addition, in a comparison experiment with ChatGPT, 64.1% of the respondents answered that the tour routes generated by the system were better. The experimental results suggest that this system has the potential to significantly improve the user's tour experience.