Information recommendation method, device and equipment

Abstract

Embodiments of the invention disclose an information recommendation method, device and equipment. The method comprises the steps of obtaining entity information contained in to-be-recommended information and association relationship information between entities corresponding to the entity information; constructing a knowledge graph based on the entity information and the association relationship information, the knowledge graph including an association relationship between an entity node corresponding to the entity information and an entity node corresponding to the association relationship information; and performing feature sampling on a target entity node of which the node popularity is greater than a predetermined popularity threshold in the knowledge graph, and determining informationneeding to be recommended to a user in the to-be-recommended information based on features sampled by the target entity node.

Description

technical field [0001] This specification relates to the field of computer technology, and in particular to an information recommendation method, device and equipment. Background technique [0002] Now people receive all kinds of recommended information almost every day, from news, shopping to catering and entertainment, etc. The recommended information is completed by the recommendation system, which is an important means of information filtering , can recommend suitable services for the user according to the user's habits and hobbies. [0003] With the rapid development of Internet technology and industry, the number of servers and webpages connected to the Internet has also increased exponentially. In the face of massive information, users can only present the same item sorting results through traditional search algorithms, and cannot target different users. Provide corresponding services according to your interests and hobbies. The above recommendation system only uses...

AI Technical Summary

Problems solved by technology

[0003] With the rapid development of Internet technology and industry, the number of servers and webpages connected to the Internet has also increased exponentially. In the face of massive information, users can only present the same item sorting results through traditional search algorithms, and cannot target different users. Provide corresponding services based on interests and hobbies

The above recommendation system only uses the historical interaction information of users and items. For new users or new items, since there is no historical interaction information related to the two in the recommendation system, accurate information recommendation cannot be made.

Images