Application recommendation method and device

Abstract

The invention discloses an application recommendation method and device. The method comprises obtaining all user web surfing records in preset time, and calculating access information of all user access applications; sorting all user access information on account of all applications and calculating all user application access interest scores based on preset sorting scoring rules; analyzing all application access interest scores of all users to obtain user categories and the similarity among all users in the user categories, and performing application recommendation based on the similarity among the users. By means of the method and the device, the access information is analyzed to obtain the interest scores of the users towards the applications, the similarity among the users is obtained through the application access interest scores of the users, and the application recommendation is achieved through the similarity between the interest scores and the users, accordingly, application recommendation is achieved, human interventions in application recommendation are prevented, and recommendation of applications interested by users is achieved.

Description

technical field [0001] This application relates to mobile Internet application technology, in particular to a method and device for implementing application recommendation based on user surfing behavior. Background technique [0002] With the rapid development of the mobile Internet, mobile terminal applications are becoming more and more abundant. According to statistics, the number of applications in the Apple App Store has reached one million, and Android applications are growing faster, most of which are web-based applications. [0003] At present, the application store mainly recommends applications to users based on the number of application downloads and user reviews. Although application recommendations can be provided to users based on information such as the number of downloads and user reviews of applications, human intervention often exists in the ranking of download times and user reviews, which affects the effect of recommending applications to users. In addit...

AI Technical Summary

Problems solved by technology

Although application recommendations can be provided to users based on information such as the number of downloads and user reviews of the application, human intervention often exists in the ranking of download times and user reviews, which affects the effect of recommending applications to users.

In addition, for each user, the business he is interested in does not necessarily match the number of downloads and user reviews, which obviously will make the application recommendation effect worse.

Since the recommendation effect does not meet the user's requirements, users often need to query and browse a large number of applications according to their own interests and needs in order to obtain the applications they need. It is impossible to complete the application acquisition based on application recommendations, which takes a long time and has a relatively poor experience. Difference

[0004] As mentioned above, the existing application recommendation methods, on the one hand, there is human intervention in application recommendation, which affects the accuracy of application recommendation; If it meets the user's interests, the effect of application recommendation is poor. Since the recommendation effect cannot meet the user's requirements, users often need to query and browse a large number of applications according to their own interests and needs, and it takes a long time to obtain the desired application.

Images