Incorporating contextual information into personalized mobile applications recommendation

With the rise of mobile Internet, number applications (apps) has shown explosive growth, which directly leads to apps data overload. Currently, recommender system become most effective method solve app However, existing recommended methods for ignore functional exclusiveness features, makes it difficult further improve recommendation performance. To this problem, we propose a personalized context-aware approach, called PCMARA. Specifically, (1) PCMARA explores relationship between contextual information and function constructs factors represent app. (2) leverages factor design novel similarity model, enable effectively eliminate adverse effects that exclusiveness. (3) comprehensively considers users generate list based on target users’ current time location. We applied real-world dataset conducted large-scale effect experiment. The experimental results demonstrate superiority PMARA over benchmark methods.