Recommendation method and device based on artificial intelligence, electronic equipment and storage medium

Abstract

The invention provides a recommendation method and device based on artificial intelligence, electronic equipment and a storage medium. The method comprises the steps of obtaining object features of ato-be-recommended object and candidate recommendation information features of each piece of candidate recommendation information, and combining the object features with the candidate recommendation information features of each piece of candidate recommendation information to form fusion features corresponding to each piece of candidate recommendation information; performing multi-level mapping processing on the fusion feature corresponding to each piece of candidate recommendation information to obtain scores of each piece of candidate recommendation information corresponding to a plurality ofindexes; performing multi-index aggregation processing on the scores corresponding to the plurality of indexes to obtain a comprehensive score of each piece of candidate recommendation information soas to perform descending sorting on the plurality of pieces of candidate recommendation information; and selecting at least one piece of candidate recommendation information sorted in the top from adescending sorting result to execute a recommendation operation corresponding to the to-be-recommended object. According to the invention, accurate recommendation of information can be realized.

Description

technical field [0001] The present invention relates to artificial intelligence technology, in particular to an artificial intelligence-based recommendation method, device, electronic equipment and storage medium. Background technique [0002] Artificial Intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. [0003] With the development of information technology and the Internet industry, information overload has become a challenge for people to process information. Personalized recommendation systems can effectively alleviate such problems by understanding user intentions more accurately. The way the ranking model predicts the click-through rate will still recommend to the user the top-ranked information that is not really liked by the user. Co...

AI Technical Summary

Problems solved by technology

[0003]With the development of information technology and the Internet industry, information overload has become a challenge for people to process information. The personalized recommendation system effectively alleviates this problem by understanding user intentions more accurately. However, the method of predicting the click-through rate through the linear sorting model or the deep sorting model in related technologies will still recommend to the user information that is ranked first but is not really liked by the user.

Images