Method and system for mining associated users in social network

Abstract

The invention discloses a mining method and system for associated users in social networks, and the method and system can improve the mining accuracy of associated users, the recall rate and the operation efficiency in a big data environment. The method comprises the steps: calculating the to-be-integrated similarity of the attributes of users in a first social network and a second social network,carrying out the integration of the similarity of the user attributes, and obtaining a user attribute similarity matrix; forming a user relation integration matrix of the first and second social networks through the tensor product; converting the user attribute similarity matrix into a user attribute similarity vector through a vectorization operator, and constructing an associated user mining model through the integration of the user relation integration matrix and the user attribute similarity matrix; and determining the associated users of the first and second social networks through solving the associated user mining model.

Description

technical field [0001] The invention relates to the field of associated user mining, in particular to a social network associated user mining method and system. Background technique [0002] Large-scale social networks are the core object of current and future social network research. The fusion of large social networks is of great value and significance to various researches on social networks. Linked user mining for large social networks is an important research content of social network integration. The current research progress of associated user mining for large-scale social network integration can be summarized as follows: [0003] (1) Mining associated users from user attributes (including user behavior) is the most researched and effective method at present. However, in a large social network, due to the similarity, sparsity, falsehood, and inconsistency of user attributes, the method of mining associated users using user attributes alone is not robust enough, and...

AI Technical Summary

Problems solved by technology

In similar or identical graphs, this method can more accurately mine nodes with higher node degrees, but it is less feasible to directly apply to associated user mining

In addition, in the current research on associated user mining based on user relationships, most mining methods require seed users; however, in many social networks, it is becoming more and more difficult to obtain seed users; currently, modeling for associated user mining without seed users There is still very little research;

[0005] (3) There are few studies on the fusion of user attributes and user relationships to mine accurate and comprehensive associated users;

[0006] (4) For large-scale social networks, there are few studies on establishing fast associated user mining modeling

Images