An Android App Recommendation Approach by Merging Network Traffic Cost

Abstract

A large amount and different types of mobile applications are being offered to end users via app markets. Existing mobile app markets generally recommend the most popular mobile apps to mobile users for purpose of facilitate the proper selection of mobile apps. However, these apps normally generate network traffic, which will consumes user mobile data plan and may even cause potential security issues. Therefore, more and more mobile users are hesitant or even reluctant to use the mobile apps that are recommended by the mobile app markets. To fill this crucial gap, we propose a mobile app recommendation approach which can provide app recommendations by considering both the app popularity and their traffic cost. To achieve this goal, we first estimate app network traffic score based on bipartite graph. Then, we propose a flexible approach based on Benefit-Cost analysis, which can recommend apps by maintaining a balance between the app popularity and the traffic cost concern. Finally, we evaluate our approach with extensive experiments on a large-scale data set collected from Google Play. The experimental results clearly validate the effectiveness and efficiency of our approach.