Knowledge recommendation method and device, electronic equipment and readable storage medium

Abstract

The invention relates to an artificial intelligence technology, and discloses a knowledge recommendation method, which comprises the following steps: counting the occurrence times of historical knowledge tags in historical access knowledge information of a user to obtain tag access weights; sorting and combining the historical knowledge tags according to the tag access weights to obtain a historical knowledge tag sequence; obtaining a to-be-pushed knowledge set containing knowledge labels; calculating the similarity between the user initial tag and the knowledge tag to obtain a first recommendation score; calculating the similarity between historical knowledge tags and knowledge tags in the historical knowledge tag sequence to obtain a second recommendation score; calculating a target recommendation score according to the first recommendation score and the second recommendation score; and screening out the to-be-pushed knowledge from the to-be-pushed knowledge set according to the target recommendation score, and sending the to-be-pushed knowledge to the user. The invention further relates to a block chain technology, and historical knowledge tags can be stored in block chain nodes. The invention further provides a knowledge recommendation device, equipment and a medium. According to the invention, the accuracy of knowledge recommendation can be improved.

Description

technical field [0001] The present invention relates to artificial intelligence technology, in particular to a knowledge recommendation method, device, electronic equipment and readable storage medium. Background technique [0002] At present, a common method of targeted push messages is to find knowledge related to or interested in users from a large amount of knowledge, so as to achieve thousands of faces, improve the hit rate of recommended knowledge, and convert them into effective clicks for users . [0003] The current knowledge recommendation method labels the knowledge (for example: real estate, finance, bonds and other labels), and then matches the corresponding knowledge to recommend according to the fixed user initialization label (such as real estate). For example, the user initialization label is "real estate When a user with the "label" recommends knowledge, it will match the knowledge corresponding to the knowledge label related to the "real estate" label. ...

AI Technical Summary

Problems solved by technology

[0004] However, in the current knowledge recommendation method, the user's label is fixed, and it cannot be adjusted when there is a deviation in the use process, resulting in low accuracy of knowledge recommendation.

Images