Socialized recommendation method based on matrix decomposition and network embedding joint model

Abstract

The invention provides a socialized recommendation method based on a matrix decomposition and network embedding joint model. The method comprises the following steps: constructing a user-article scoring matrix and user-user social network and generating a user social corpus according to user-user social network; utilizing a user-article scoring data and a user social corpus training matrix decomposition and network embedding joint model to obtain a user feature matrix and an article feature matrix; predicting an unobserved score according to the user feature matrix and the article feature matrix; and recommending a plurality of articles with relatively high prediction score values to the corresponding users. According to the method, a matrix decomposition model and a network embedding model are seamlessly integrated by designing a unified target function; based on a unified optimization framework, bidirectional promotion and collaborative optimization between a matrix decomposition model and a network embedding model are realized, so that interested articles can be accurately recommended to a user.

Description

technical field [0001] The invention relates to the technical field of computer applications, in particular to a social recommendation method based on a matrix decomposition and network embedding joint model. Background technique [0002] In the Internet age, how to effectively adjust the contradiction between the richness of diversified information and the limitation of people's attention has become a technical problem that the current information industry needs to solve urgently. In this context, the recommendation system came into being to deal with the information overload problem caused by big data. [0003] Collaborative filtering is the core technology behind the recommendation system, which predicts the unobserved "user-item" correlation by analyzing the "user-item" interaction history. The matrix factorization model is currently the most mainstream collaborative filtering method. The core idea of ​​matrix factorization is: by decomposing the "user-item" interaction...

AI Technical Summary

Problems solved by technology

[0007] 1) Conventional network embedding models use unsupervised learning methods, and their purpose is general rather than customized for recommendation tasks; while social networks are complex and multifaceted, it is difficult for network embedding models to mine those Social properties that contribute to recommender systems

[0008] 2) Due to the separate two-stage design, the objective functions of the network embedding model and the matrix factorization model are not unified, so that the optimal results generated in the first stage may not be optimal for the second stage of recommendation tasks; in addition, the network embedding model Contains a large number of parameters, making it very difficult to jointly tune the two models

Images