Content recommendation method and device

Abstract

The invention discloses a content recommendation method and device, and relates to the field of content recommendation. According to the method, multiple items of to-be-classified content in a contentpool are obtained and content identification is carried out; a corresponding content classification model is selected according to a content identification result to perform content classification onthe to-be-classified content to obtain to-be-recalled content; and preliminary recall is performed on the to-be-recalled content according to a recall strategy to obtain recalled content, the recalled content is sorted according to a recommendation model to obtain a preliminary sorting list, secondary sorting is performed on the preliminary sorting list by using a sorting algorithm model, and integration is performed to obtain a recommended content list. The content is recognized, filtered and classified accurately before being published by controlling the source, normal content is exposed tothe user online, and low-quality content is screened and filtered and is not exposed to the user so that online of low-quality content such as sensational headline writers, taking advantage of loopholes or vulgar on the platform is avoided, the overall content quality of the platform is improved and the stickiness of platform users is kept.

Description

technical field [0001] The present invention relates to the field of content recommendation, in particular to a content recommendation method and device. Background technique [0002] Nowadays, in many Internet products, especially content platforms, the content recommendation system is an indispensable part, which can provide users with high-quality personalized recommendation services without explicit behavior of users. For example, when a user opens a news APP, content and articles of interest to the user appear on the homepage. To meet the needs of the user, it is necessary to use some novelty recommendation results to play the role of traffic distribution, and at the same time, it is necessary to provide users with accurate and personalized recommendations, thereby shortening the user's product list. selection period. This puts forward higher requirements for the diversity and accuracy of the recommendation system. How to attract users to pay attention to the pushed c...

AI Technical Summary

Problems solved by technology

[0003] However, in order to attract traffic, the publishers of some articles publish partisan, edge-cut or vulgar content. These content attract users to click and get more exposure and recommendations. Although users click to read the content, they don’t get much. A good user experience will lead to vulgar content on the platform in the long run, resulting in a large loss of users. Therefore, it is necessary to propose a method that can filter and classify the content published on the platform, remove low-quality content such as vulgar and headlines, and improve the quality and quality of user recommended content. Content Recommendation Method for Conversion Rate to Improve the Accuracy of Content Recommendation

Images