Matrix decomposition project recommended algorithm for social status enhancement of users

Abstract

The invention discloses a matrix decomposition project recommended algorithm for social status enhancement of users. Firstly, user project scoring matrixes are divided according to project categories;user scoring and the principle of the social relationship co-occurrence are used to infer specific category user social network; PageRank algorithm is used to calculate the social status of the userson the inferred specific category user social network; the scoring weights of the users are measured with social status values of the users, the specific category user scoring data and the social relationship data are combined to carry out the matrix decomposition, implicit expression feature vector quantity of the users and project in specific field are learned, the implicit expression feature vector quantity inner product of the users and project are used to predict the scoring of the users to the project, and the personalized project recommendation is provided for the users. The matrix decomposition project recommended algorithm for social status enhancement of users efficiently solves the two following problems, which is traditional and neglected based on social network recommendation technology: (1) in different fields, the users trust different friends; (2) because the users have different social status in different field, the users have different influence from friends in different fields.

Description

technical field [0001] The invention belongs to the technical field of data mining, and in particular relates to a matrix decomposition item recommendation method for enhancing user social status. Background technique [0002] With the continuous development of Internet technology, it is becoming more and more difficult to find valuable relevant information from massive data. The recommendation system analyzes the user's historical activity data, mines the user's potential preferences, and provides users with personalized recommendation services. It has become an effective means to solve the problem of information overload, and has attracted extensive attention from academia and industry in recent years. In the research of recommendation system, collaborative filtering algorithm is the most widely used recommendation technology. The collaborative filtering algorithm predicts the user's future preferences by analyzing the user's historical feedback information. However, col...

AI Technical Summary

Problems solved by technology

However, the existing recommendation algorithms based on social networks ignore the following two facts: (1) in different fields, users trust different friends; (2) users are not only influenced by different friends in different fields, but also different Users are influenced differently by friends

Images