Utility Mining-based Point-of-Interest Paths Recommendation Using SNS Posts in Pervasive Social Environments

Abstract

In this paper, we propose a point-of-interest (POI) paths recommendation method for SNS-savvy travelers, called a Utility Mining-based POI Path Recommendation (UMPOIR) framework. To recommend the best POI paths depending on the traveler’s context, UMPOIR framework considers the popularity and the personal preferences for the POIs. Furthermore, it reflects not only the current context of the traveler, but also the experience of the publishers who visited there. So, it calculates the utility of the candidate POIs and their paths. After this, the utility is refined by the popularity of the POIs, the traveler’s personal preferences, and the temporal factors. To evaluate the performance, we conducted experiments to measure the adequacy and the degree of satisfaction. We proved that UMPOIR framework can improve the recommendations’ adequacy and the degree of satisfaction with the recommendations.