Audio-visual recommendation system and method based on bilinear perceptual graph neural network model

Abstract

The present invention provides an audio-visual recommendation system and method based on a bilinear perceptual graph neural network model, which calculates the similarity between the obtained program vector and the pre-acquired user embedding vector through the similarity function; and then uses the predictive activation function (sigmoid activation function ) to limit the range of similarity to the preset interval, so that the value in the preset interval can be used as the possibility evaluation score of the user's audio-visual choice, and it is constructed as a personalized audio-visual recommendation model based on the knowledge map - a dual Linear knowledge-aware graph neural network model system; the system captures the second-order feature interaction information of neighbor nodes in the knowledge graph by adding a bilinear collector on the basis of the graph neural network, obtains the program knowledge vector, and expresses it through the program text The module obtains the program text vector, and then obtains the recommendation data based on the program knowledge vector and the program text vector. In this way, the accuracy of audio-visual recommendation is improved, the operation effect of audio-visual recommendation is improved, and user stickiness is improved.

Description

technical field [0001] The present invention relates to the technical field of media recommendation, and more specifically, to an audio-visual recommendation system based on a bilinear perceptual graph neural network model. Background technique [0002] With the development of the Internet and the advent of the era of information explosion, people have completed the transition from information scarcity to information overload. The emergence of personalized recommendation system relieves the pressure of information overload and helps users obtain information that is really helpful to them from massive amounts of data. Therefore, the knowledge map that can be used as auxiliary information can be introduced into the recommendation system to effectively alleviate the data sparsity and cold start problems suffered by the collaborative filtering algorithm and improve the recommendation effect. Therefore, graph neural network technology can be used to capture and learn from it. re...

AI Technical Summary

Problems solved by technology

[0005] In view of the above problems, the purpose of the present invention is to provide an audio-visual recommendation system based on a bilinear perceptual graph neural network model to solve the problem in the prior art that the recommendation algorithm based on a graph neural network is usually implemented using a GNN+linear collector Propagation and aggregation of neighbor information, that is, by superimposing the neighbor information of a node on its own information in a linear manner, and performing the above operations multiple times, although this method can extract the features of the graph data and capture the high-order features between features connectivity, but ignores the feature interaction information between two adjacent nodes; in addition, text data (such as IPTV movie program titles and program introductions, etc.) are generally not widely used in recommendation algorithms. As a result, the current recommendation system lacks the proportion of text-based information, and text-based data contains rich information and has a greater impact on user preferences when using it. Therefore, the current recommendation method has low recommendation accuracy and poor adaptability.

Images