Predicting POI visits with a heterogeneous information network

A point of interest (POI) is a specific location that people may find useful or interesting. Examples include restaurants, stores, attractions, and hotels. With recent proliferation of location-based social networks (LBSNs), numerous users are gathered to share information on various POIs and to interact with each other. POI recommendation is then a crucial issue because it not only helps users to explore potential places but also gives LBSN providers a chance to post POI advertisements. As we utilize a heterogeneous information network to represent a LBSN in this work, POI recommendation is remodeled as a link prediction problem, which is significant in the field of social network analysis. Moreover, we propose to utilize the meta-path-based approach to extract implicit (but potentially useful) relationships between a user and a POI. Then, the extracted topological features are used to construct a prediction model with appropriate data classification techniques. In our experimental studies, the Yelp dataset is utilized as our testbed for performance evaluation purposes. Results of the experiments show that our prediction model is of good prediction quality in practical applications.