Abstract
k-Clique detection enables computer scientists and sociologists to analyze social networks' latent structure and thus understand their structural and functional properties. However, the existing k-clique-detection approaches are not applicable to signed social networks directly because of positive and negative links. The authors' approach to detecting k-balanced trusted cliques in such networks bases the detection algorithm on formal context analysis. It constructs formal contexts using the modified adjacency matrix after converting a signed social network into an unweighted one. Experimental results demonstrate that their algorithm can efficiently identify the trusted cliques.
Original language | English (US) |
---|---|
Article number | 6777472 |
Pages (from-to) | 24-31 |
Number of pages | 8 |
Journal | IEEE Internet Computing |
Volume | 18 |
Issue number | 2 |
DOIs | |
State | Published - Mar 1 2014 |
Keywords
- FCA
- equiconcept
- signed social networks
- trusted cliques
ASJC Scopus subject areas
- Computer Networks and Communications