User preference prediction method based on multivariate credit evaluation

Abstract

The invention provides a user preference prediction method based on multivariate credit assessment, and belongs to the technical field of internet user preference prediction. The method comprises thefollowing steps: firstly, constructing a trust adjacency matrix between a target user and other users and a score sparse matrix between the target user and a commodity; obtaining the comprehensive trust degree and importance degree of the target user according to the trust adjacency matrix and the score sparse matrix, and performing normalization processing on the comprehensive trust degree and importance degree of the target user; according to the comprehensive trust degree and importance degree after normalization processing, establishing a scoring prediction model of the target user for thecommodity, constructing an objective function of the scoring prediction model, optimizing the objective function through a gradient descent method, and obtaining a user preference vector and a commodity feature vector; and calculating the target user preference according to the user preference vector and the commodity feature vector. According to the method, the direct trust relationship betweenthe users and the indirect trust relationship transmitted by the intermediary user are considered, the influence of malicious users on preference prediction is eliminated, the calculation complexity is low, and prediction is more accurate.

Description

technical field [0001] The invention relates to the technical field of Internet user preference prediction, in particular to a user preference prediction method based on multiple credit evaluations with more accurate prediction. Background technique [0002] The emergence and popularization of Web 2.0 has brought a large amount of information to Internet users, which has satisfied users' needs for information in the information age. When faced with a large amount of information, it is impossible to obtain the information of interest and useful information to oneself, which reduces the efficiency of information use. This is the so-called information overload problem. [0003] A very potential solution to the problem of information overload is the recommendation system, which is a personalized information recommendation system that recommends the information and products that the user is interested in to the user according to the user's information needs and interests. Compar...

AI Technical Summary

Problems solved by technology

The sparsity of trust information leads to inaccurate calculation of trust value between users and inaccurate calculation of user influence, which leads to inaccurate prediction of user preference

[0010] Third, the existing recommendation system does not take into account the distrust relationship between users, resulting in the failure to find malicious ratings / comments of malicious users, which affects the calculation accuracy of trust values ​​and damages the performance of the recommendation system

Images