Personalized recommendation method combining trust and influence based on deep learning

Abstract

The invention discloses a personalized recommendation method combining trust and influence based on deep learning. The method specifically comprises the following steps: constructing a user-project scoring matrix; performing feature extraction on the data; calculating the user-project scored matrix to obtain the initial direct trust degree of the user, calculating the user similarity by using Pearson correlation coefficient measurement, and obtaining the preference degree of the user to the project in combination with the user interaction frequency; distributing different weights to differentprojects according to preference degrees of users to the projects in successful or failed user interaction to obtain final direct trust degrees, and forming a user-user trust matrix; mining user nodeswith relatively large influence in the social network by utilizing topological structure information in the trusted network; constructing and training a model; and predicting scores of unknown projects by the user through the trained model, and selecting projects with higher scores to generate a recommendation set. More accurate feature vectors are obtained by using deep learning, so that the recommendation accuracy is improved.

Description

technical field [0001] The present invention relates to the field of recommendation system and social network, in particular, to a personalized recommendation method based on deep learning combining trust and influence. Background technique [0002] With the development of the information age, the increasingly large data flow on the Internet makes it more and more difficult for people to obtain the information they need, and information overload has become an urgent problem to be solved. The information filtering technology that helps us filter useful data from massive data is becoming more and more important. The recommendation system is an ideal method to find data of interest to users from large-scale data according to user preferences. But in the increasingly complex social network environment, the sparsity of user-item rating matrix and the weak transfer of trust still affect the recommendation accuracy. Improving the accuracy and performance of the system has become a...

AI Technical Summary

Problems solved by technology

However, although the collaborative filtering algorithm can improve the recommendation accuracy to a certain extent, it faces the problems of "data sparsity" and "cold start" in practical applications.

The "data sparse" problem refers to the problem that there are too many empty elements in the user-item matrix and too few valuable elements, resulting in too little available data; matter of preference

Images