Graph neural network link prediction recommendation method based on social relations

Abstract

The invention discloses a graph neural network link prediction recommendation method based on social relations, which comprises the following steps: establishing a user-article bipartite graph, and constructing an article interaction topological graph containing the social relations of users according to the social relations among the users. Taking the comment text and the purchase relationship assources of node information, and performing feature extraction work of text data and a network structure by using Bert to obtain an initial feature vector of each node; by using a residual connectionmode on the GNN, enabling the node information in the graph to reserve the structure information of the original graph as much as possible, and obtaining the updated node information. And finally, obtaining the preference degree of the user to the article through a link prediction algorithm, and generating a recommended article list by adopting Topn recommendation for the obtained prediction score. According to the method, node features are subjected to personalized description through comments, and structure information of a topological graph formed by users, namely articles, can be fully utilized, so that more effective recommendation is carried out.

Description

technical field [0001] The invention relates to the technical fields of recommendation system, text classification and GNN research, in particular to a graph neural network link prediction and recommendation method based on social relations. Background technique [0002] With the continuous development of Internet technology and the sharp increase in the amount of online information, it is often difficult for users to choose products that really suit them. Therefore, the demand for personalized recommendation services is becoming more and more extensive. How to make full use of users' existing comments and behavior data, mine user preferences and product characteristics, and recommend suitable products for users is very important for improving user experience and increasing product sales. Therefore, the recommendation system has become a hot field in the current Internet development process. [0003] Traditional collaborative filtering recommendation technology generally us...

AI Technical Summary

Problems solved by technology

[0005] (1) Only relying on a single rating data, the review information is not fully utilized;

[0006] (2) User social information is rarely used

Images