Network community detection adversarial enhancement method based on multi-similarity integration

Abstract

The invention discloses a graph network community detection adversarial enhancement method based on similarity. The method comprises the following steps: S1, loading a network; S2, performing networkreconnection based on similarity; S3, performing community detection to acquire community division; S4, integrating a plurality of community division results, and constructing a new community structure, which comprises the following specific operation steps: 4.1) constructing a co-occurrence network; 4.2) selecting an optimal threshold; 4.3) performing network pruning; and 4.4) performing isolatenode distribution; according to the network community detection adversarial enhancement method, the network structure is repaired and enhanced through the similarity-based link prediction, multiple community divisions are aggregated through integrated clustering, the stability of the network community structure can be improved, and the performance of a community detection algorithm is greatly improved.

Description

technical field [0001] The invention relates to the fields of network science and data mining, in particular to a network community detection confrontation enhancement method based on multi-similarity integration. Background technique [0002] Nodes with high similarity in the network are aggregated into subgraphs, which are called communities. The connections within the communities are dense, and the connections between communities are sparse. For example, in a social network, a community often reflects a collection of users with the same topic of concern; in a collaborative network of scientists, a community is composed of scholars with the same research field. The community structure of the network contains important topological features of the network. A large number of studies have shown that there is a large difference between the characteristics of the community level and the global characteristics in the network, and ignoring the community structure of the network w...

AI Technical Summary

Problems solved by technology

[0005] Aiming at these difficulties in the prior art, the present invention proposes the concept of network community detection and confrontation enhancement. The present invention aims to solve the following problems by combining link prediction and integrated clustering: repair and enhance network connections through link prediction based on similarity indicators, Make the originally inconspicuous or damaged network community structure clear and stable, and use integrated clustering to aggregate multiple community divisions to obtain a more accurate community structure, thereby helping existing community detection algorithms to improve detection accuracy and better application Discover tasks in online communities

Images