article · Procedia Computer Science
Knowledge extraction from social networks is increasingly useful across political, socio-economic, and scientific domains, with community detection serving as a primary mechanism. Existing algorithms often struggle to remain constant, effective, and accurate when processing networks that feature high volumes of edges. To address this limitation, a hybrid community detection approach named PercoMCV was developed for networks exhibiting numerous links between communities. The method operates in two distinct stages. First, it identifies potential communities using a clique percolation algorithm. Second, it applies the Eigenvector Centrality method to these results to evaluate the influence of individual nodes, thereby reducing the proportion of unclassified nodes across the network. Testing across various network types showed that the method effectively uncovers relevant communities and outperforms alternative community detection algorithms in performance and accuracy.
Social networks influence modern life in areas ranging from politics to science, but mapping their complex structures remains difficult. By improving how communities are grouped within heavily connected networks, this approach helps analysts accurately classify individuals and identify influential points within large, intricate groups of data.
The method could be used by data analysts or software developers building network analysis tools to improve community discovery in densely connected datasets. Because the work is validated on test networks within an academic study, it appears to be early-stage research requiring further development and integration before commercial deployment.
AI-generated from the published abstract. Always read the original work before citing.
Knowledge extraction in social networks is a needful tool as it touches every aspect of our lives such as politic, socio-economic, scientific, etc. Community detection is one of the objectives of this specific tool used for knowledge extraction in social networks. Many algorithms of knowledge extraction from social networks have been developed these last years. However, many of them are not constant, effective and accurate when facing these social networks with many edges. In this paper, we propose a new approach of community detection in social networks with many links between communities. The proposed approach has two steps. In the first step, the algorithm attempts to determine all communities that the clique percolation algorithm may find. In the second step, the algorithm computes the Eigenvector Centrality method on the output of the first step in order to measure the influence of network nodes and reduce the rate of the unclassified nodes. To assess this new approach, we test it on different types of networks. Relevant communities that have been detected testifies effectiveness and performance of the approach over other community detection algorithms.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1016/j.procs.2019.04.010
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
