In this paper, we propose a modified Cluster Walk Trap (CWT) approach combined with Closed Cycle Approach(CCA). The proposed method uses probability measure in a multimode network to evaluate group of persons involved in common crime and group of crime done by a common person thereby avoiding the difficulty of adjacency matrix. The usage of transitivity based on closed cycle in node identification overcomes the problem of computational complexity with the increased number of nodes and edges. Based on the clustering coefficient value, the influential nodes are classified as victim, suspect and witness. Experiments are conducted using the publicly available KONECT dataset and the simulation results show that the proposed approach is good in identification of communities and influential nodes with improved degree of accuracy than the other approaches reported in the literature.
Digital Object Identifier (DOI)
Pugalendhi, Ganeshkumar and Kumaresan, Shanmugapriya
"Customized CWT-CCA: Discovery of Prominent Persons in the Crime Network,"
Applied Mathematics & Information Sciences: Vol. 13:
4, Article 18.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol13/iss4/18