Recommendation system score prediction method based on graph neural network and attention mechanism

Abstract

The invention discloses a recommendation system score prediction method based on a graph neural network and an attention mechanism, and the method comprises the following steps: S1, converting a userproject score graph into a user project score credibility graph based on degree and time information, and sampling neighbor vertexes for each vertex in the graph; s2, updating the state of each vertexin the user project scoring credibility graph by utilizing a sampling result and combining time information; s3, performing score prediction on the project by the user based on an attention mechanism, and updating a score prediction model; and S4, using the score prediction model to realize score prediction of the project by the user. According to the invention, an application way of the graph neural network in the universal recommendation system is given, static characteristics of users and projects are combined, the graph representation learning ability of the graph neural network is utilized to learn the importance degree of hidden characteristics in the user project interaction network, and more attention is paid to serve the recommendation system.

Description

Technical field[0001]The present invention belongs to the technical field of recommendation system, and more particularly to a recommended system score prediction method based on map neural network and attention mechanism.Background technique[0002]The recommendation system is designed to recommend a potential association project for users, with projects that refer to all kinds of resources available for movies, music, web pages, and commodities. In the recommended system, the user's scoring premise for the project is premise for recommendation and ranking, which is also described as a matrix fill, ie, filling the score data in the score matrix, and the traditional diagram represents a learning method, based on matrix Techniques and other technologies generate or learn to obtain the vector representation of the user and the project, which in turn utilizes vector and dot coordinating operation to implement rating prediction. Such scoring prediction methods focus on excavating static u...

AI Technical Summary

Problems solved by technology

This kind of rating prediction method focuses on mining the static user-item interaction network structure features, and there are three problems: first, the static characteristics of user vertices and item vertices are not fully utilized, and the accuracy of rating prediction can easily reach the bottleneck; second , time information is seldom considered, and the characteristics of insensitivity to time make this type of model unable to capture the evolution characteristics of the network, and it is difficult to ensure the long-term effectiveness of the recommendation process based on predicted ratings; item representation learning process, but neglects to measure the importance of user-item representations in the rating prediction task

Images