A content recommendation method, device, electronic device and storage medium

Abstract

The present application relates to the field of computer technology, and in particular to a content recommendation method, device, electronic device and storage medium, which can be applied to scenarios such as cloud technology, artificial intelligence, intelligent transportation, and assisted driving to improve the timeliness of the recommendation system. The method includes: responding to the content recommendation request, based on the existing duration of the candidate content, selecting the content to be recommended from the candidate content set; obtaining the object characteristics of the target object, the context characteristics related to the content recommendation request, and the content characteristics of each content to be recommended, The content features include: based on at least one of the content release time and the occurrence time of content-related events, the existence time feature that characterizes the existing time of the content; based on the above features, low-level feature crossover and high-level The first order feature crosses to determine the target recommendation value; the recommendation order is determined based on the target recommendation value. This application recalls and sorts the content based on the existing length of time, which can effectively improve the timeliness.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a content recommendation method, device, electronic equipment and storage medium. Background technique [0002] With the rapid development of computer technology, more and more content needs to be obtained through computer processing, and then transmitted through the network after processing; and more and more scenarios need to recommend content to users. For example, recommendations for news, videos, or advertisements. [0003] In related technologies, content recommendation mainly relies on content recommendation models. In order to ensure that time-sensitive content is ranked first, two weighting methods are proposed in the related art. [0004] The first way is: weighting the estimated scores of time-sensitive items estimated based on the content recommendation model. [0005] When using the first method, the weighted value will be coupled with the content ...

AI Technical Summary

Problems solved by technology

[0007] When using the second method, because the weighting of high-timeliness samples has no causal relationship with the improvement of their estimated scores, the weighting of samples cannot effectively improve the corresponding estimated scores of high-timeliness samples, thereby affecting the ranking results

[0008] In summary, neither of the above two weighting methods can effectively improve the timeliness of the recommendation system

Images