MR-UI: A Mobile Application Recommendation Based on User Interaction

With the rapid growth of mobile applications in major app stores, it is hard for users to choose their desired mobile applications. Therefore, it is necessary to provide a high-quality mobile application recommendation mechanism to meet the user's expectation. However, the existing recommendation methods are still not accurate enough in the embedding representations of users and mobile applications. Based on the neural graph collaborative filtering technique, we propose a mobile application recommendation method based on user interaction to solve this problem. First of all, by introducing the high-order connectivity between users and mobile applications, it exploits the embedding propagation to capture the collaborative filtering signals along the graph structure to further refine the embedding representations between mobile applications and users. Then, the user preferences for different mobile applications are predicted through inner product, and the recommendation task is completed. The real dataset of Kaggle is used to evaluate our approach and the experimental results show that our recommendation method can achieve the best results in different evaluation metrics. It can effectively improve the recommendation accuracy for mobile applications.