Personalized Recommender System for Smartphones based on Application Usage

Abstract: The rebellious growth of the mobile application market has made it a significant challenge for the users to find relevant applications in crowded Application Stores. To diminish this problem, existing solutions often use the user's application-download history or user-rating to recommend applications that might interest them. However, the user downloading an application does not indicate that the user likes that application. Using user-ratings, on the other hand, suffers from tedious manual input and potential data insufficiency problems. In this paper, we present a system that makes personalized application recommendations by analyzing how the user actually uses his installed applications. Based on all user's application usage records, our system employs an item-based collective filtering algorithm for individualized recommendations.

Keywords: mobile application market, user rating, recommend applications, item-based collective filtering algorithm.