Detecting k-Balanced Trusted Cliques in Signed Social Networks
IEEE Internet Computing
Institute of Electrical and Electronics Engineers (IEEE)
This is the author accepted manuscript. The final version is available from Institute of Electrical and Electronics Engineers (IEEE) via the DOI in this record.
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.
Vol. 18, Iss. 2, pp. 24 - 31