Itinerary recommendation using historical and interaction data

Abstract

Itinerary planning is a complex task for travelers. While some travelers have specific travel plans and preferences, there are travelers who do not have clear plans but wish to visit different points of interest. Supporting these travelers is important in a mobility society. With the aim of supporting a richer travel experience, this paper examines itinerary construction methods based on mathematical optimization and user interaction. Based on maximum likelihood programming, this paper proposes a new method that can generate a variety of itineraries as candidates by using data collected through user interaction, and reports the results of verifying its operation.