Incorporating Memory-Based Preferences and Point-of-Interest Stickiness into Recommendations in Location-Based Social Networks

In location-based social networks (LBSNs), point-of-interest (POI) recommendations facilitate access to information for people by recommending attractive locations they have not previously visited. Check-in data and various contextual factors are widely taken into consideration obtain people’s preferences regarding POIs in existing POI recommendation methods. psychological effect-based recommendations, the memory-based attenuation of with respect POIs, e.g., fact that more attention is paid were checked recently than those visited earlier, emphasized. However, memory effect only reflects changes an individual’s check-in trajectory cannot discover important dominate their mobility patterns, which related repeat-visit frequency individual at a POI. To solve this problem, paper, we developed novel framework using stickiness, named U-CF-Memory-Stickiness. First, used preference-attenuation mechanism emphasize personal effects preference evolution human patterns. Second, took visiting introduced concept stickiness identify reflect stable interests behavior decisions. Lastly, incorporated influence both user-based collaborative filtering improve performance recommendations. The results experiments conducted on real LBSN dataset demonstrated our method outperformed other