Object recommendation method and device based on artificial intelligence and electronic device

Abstract

The invention provides an object recommendation method and device based on artificial intelligence, an electronic device and a computer readable storage medium. The method comprises the following steps of constructing a tree structure according to a plurality of candidate objects, wherein the tree structure comprises a plurality of layers, the plurality of layers respectively correspond to the object recommendation models, and each layer in the plurality of layers comprises at least one node; screening the plurality of nodes in the tree structure according to the similarity relationship between a training object and the plurality of candidate objects, and constructing a training sample according to the training object and the screened nodes; alternately training the object recommendation model and the tree structure according to the training sample, wherein the trained object recommendation model and the trained tree structure are used for determining a to-be-recommended object in the candidate objects. By means of the present invention, the training effect on the object recommendation model and the tree structure can be improved, and therefore the object recommendation precision is improved, for example, the multimedia data recommendation precision is improved.

Description

technical field [0001] The present application relates to artificial intelligence technology, and in particular to an artificial intelligence-based object recommendation method, device, electronic equipment, and computer-readable 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. Machine learning (Machine Learning, ML) is the core of artificial intelligence, specializing in the study of how computers simulate or implement human learning behaviors to acquire new knowledge or skills, and reorganize existing knowledge structures to continuously improve their performance. [0003] Personalized recommendation is an important application of machine learning. The goal of personalized recommendat...

AI Technical Summary

Problems solved by technology

However, the expressive power of the vector representation is very limited, leading to a large discrepancy between the obtained cosine similarity and the actual similarity, which further leads to low accuracy of object recommendation.

Images