User interest detection method, device and storage medium

Abstract

The present invention provides a user interest detection method, device and storage medium, wherein the method includes: obtaining the interest distribution of the current user; wherein, the interest distribution includes the probability value that the current user is interested in all points of interest; based on the interest distribution and the detection situation , sample a preset number of POIs; obtain the detection content corresponding to the preset number of POIs, and push the detection content to the current user’s recommendation page to realize a round of interest detection; wherein, when the current user refreshes the recommendation page, based on the current The user's feedback behavior on the detection content displayed in the recommended page before refreshing, updates the probability value of the first point of interest corresponding to the detection content in the interest distribution, and based on the correlation between the interest points, updates the interest distribution corresponding to the detection content The probability values ​​of other second interest points related to the first interest point. The invention improves the comprehensiveness and accuracy of interest detection of cold-start users, as well as the efficiency of interest detection.

Description

technical field [0001] The present invention relates to the technical field of information processing, in particular to a user interest detection method, device and storage medium. Background technique [0002] In the personalized recommendation scenario, new users are a special type of users: first, new users lack a clear understanding of the community, and do not know what kind of content there is in the community; second, the interest portraits of new users are very sparse, resulting in It is difficult for the recommendation system to guess the content that the user likes. At the same time, new user groups are the main driving force for the continued growth of the community, and their product experience and subsequent retention and activity are very important to the community. Therefore, doing a good cold start for new users has become an important and challenging task. [0003] From the perspective of recommendation technology, a core is to detect the interest of new u...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a user interest detection method, device and storage medium to solve the defects of low efficiency and insufficient accuracy in the detection process in the prior art

Images