Abstract
With the development of Location-Based Social Networks, successive Point Of Interest (POI) recommendation systems have become a hot spot in the field of recommendation systems. Successive POI recommendation systems suggest to users new and interesting places to visit. However, in real-life POI recommendation, there are often a small number of users facing a huge number of POIs. Traditional successive POI recommendation methods are not capable to deal with the large sparse datasets, as they only consider the simple relationship between users and POIs. They do not use the context information of users and POIs, which can enable better recommendation results. To utilize the context information, this paper proposes a novel successive POI recommendation method, SQPMF, which integrates user personal preferences, user social relationships and POI transition relationships into the system for accurate recommendation of the next POI. Our experimental evaluation using three real-life datasets, Gowalla, Foursquare and Brightkite, shows that our method SQPMF consistently outperforms all state-of-the-art methods in recommendation of successive POIs. Compared with other methods, SQPMF improves Precision, Recall and F1-score by an average of at least 6.1%, 5.8% and 5.7% respectively on three publicly available datasets.