Using Attributes Explicitly Reflecting User Preference in a Self-Attention Network for Next POI Recommendation

With the popularity of location-based social networks such as Weibo and Twitter, there are many records points interest (POIs) showing when where people have visited certain locations. From these records, next POI recommendation suggests that a target user might want to visit based on their check-in history current spatio-temporal context. Current methods mainly apply different deep learning models capture preferences by nonlinear relations between POIs preference pay little attention mining or using information explicitly reflects preference. In contrast, this paper proposes utilize data reflect include in learning-based process better Based self-attention network, utilizes attributes month check-ins categories during time, which indicate periodicity user’s work life can habits users. Moreover, considering distance has significant impact decision whether POI, we used filter remove candidate were more than away recommending POIs. We use from New York City (NYC) Tokyo (TKY) datasets, experiments show improvements improve recommended performance POI. Compared with state-of-the-art methods, proposed method improved recall rate 7.32% average.