Group preference queries for location-based social networks

Abstract

Location-based social networks involve a great number of POIs (points of interest) as well as users’ check-in information and their ratings on POIs. We note that users have their own preferences for POI categories. In addition, they have their own network of friends. Therefore, it is necessary to provide for a group of users (circle of friends) a new kind of POI-finding service that considers not only POI preferences of each user but also other aspects of location-based social networks such as users’ locations and POI ratings. Aiming to solve this problem, in this paper we present a new type of query called Spatial Group Preference (SGP) query. For a group of users, an SGP query returns top-k POIs that are most likely to satisfy the needs of users. Specially, we propose a new evaluation model that considers user preferences for user preferences for POI categories, POI properties including locations and ratings, and the mutual influence between POIs. Based on this model, we develop algorithms based on R-tree to evaluate SGP queries. We conduct experiments on a simulation dataset and the results suggest the efficiency of our proposal.