Method and server for recommending mobile applications

Abstract

The invention provides a recommendation method, a client and a server of mobile applications. The method comprises steps that the client records the time of a user starting the client, scans and records a mobile application list of the user so as to acquire log data of the user and then transmits the log data to the server; the server acquires use information of every mobile application according to the received log data, and based on the use information, calculates the mobile applications and acquires scores of the mobile applications, the scores including inherent scores and personalization scores; and the server, according to the scores and relative information of the mobile applications, generates a mobile application recommendation list by using collaborative filtering, and then transmits the recommendation list to the client and displays the recommendation list to the user. According to the method, the client and the server, diversity of mobile application recommendation is increased, and influence caused by human malicious scoring is prevented.

Description

technical field [0001] The invention relates to the technical field of mobile Internet, in particular to a method for recommending mobile applications, a client and a server. Background technique [0002] With the development of mobile Internet technology, mobile terminals are becoming more and more important in people's lives. By installing various applications on mobile terminals, people's lives are greatly improved. However, with the increase of the number of mobile applications, people are faced with how to find the mobile applications they are interested in among the massive mobile applications. In order to solve this problem, people have developed some mobile application management clients, which can manage the user's mobile application and recommend other mobile applications to the user. However, the existing mobile application recommendation methods have the following problems: [0003] 1. Most recommendation methods are based on the number of downloads or user rati...

AI Technical Summary

Problems solved by technology

However, for some mobile applications, there are cases of malicious scoring, that is, artificially giving high scores to some low-quality mobile applications to attract users to download them.

Recommendation methods based on user ratings cannot effectively identify such mobile applications

[0005] 3. For new mobile applications, due to the lack of user ratings, it is difficult for traditional collaborative filtering-based recommendation methods to calculate the similarity between the application and existing applications, resulting in inaccurate or ineffective recommendation of such applications For users, there is a cold start problem

Images