Social network recommendation model construction method based on hyper-graph structure

Abstract

The invention is applicable to the field of personalized recommendation and provides a social network recommendation model construction method based on a hyper-graph structure. The method comprises the following steps: constructing a hyper-graph which takes a project as the center and a hyper-graph which takes a user as the center based on a user-project scoring matrix in a social network; calculating user evaluation similarity, project characteristic similarity and user characteristic similarity; on the basis of a matrix factor decomposition model, fusing the user evaluation similarity, the project characteristic similarity and the user characteristic similarity to obtain a target function; processing the target function by adopting a random gradient descent algorithm; carrying out iteration to solve a user potential factor matrix and a project potential factor matrix; based on the user potential factor matrix and the project potential factor matrix, predicating scores of a user on a project; and recommending a project with the highest predicated score to the user. According to the social network recommendation model construction method provided by the embodiment of the invention, on the basis of the matrix factor decomposition model, user characteristics, project characteristics and user scores are fused and recommended model description is more comprehensive; and the recommendation accuracy is improved.

Description

technical field [0001] The invention belongs to the technical field of personalized recommendation, in particular to a method for constructing a social network recommendation model based on a hypergraph structure. Background technique [0002] With the exponential growth of network information, how to improve the efficiency of information utilization and alleviate the problem of information overload has always been an important research field. Among them, the recommendation system is an important way to solve the above problems. At present, the recommendation system plays an important role in the fields of e-commerce, information retrieval, smart tourism, online advertising, mobile applications, and public opinion prediction. Since Netflix announced the recommendation system competition in 2006, it has aroused It has aroused the interest of many scientific researchers, and the recommendation accuracy has become the most important measurement index of each recommendation syst...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides a method for constructing a social network recommendation model based on a hypergraph structure, which aims to solve the traditional KNN collaborative filtering recommendation. Since only neighbor user rating information is used to predict recommended items, resulting in low recommendation accuracy question

Images