An Outlier Analysis to Tag Elites in Clandestine Social Networks

The homeland security has become a very significant consideration for all the Governments throughout the world. To improve the OPerational SECurity (OPSEC), multi relational graphs were introduced for Covert Network Analysis (CNA). In this paper, proposed SpyNetMiner system identifies the key players who maximally influence the covert network. Abnormality of the nodes is analyzed based on the profile generated using enhanced selection strategies. It further justifies the findings by presenting layman understandable explanation through feature extraction and semantic rule convertors. An event that brought a worldwide attention towards terrorism is the unforgettable 9/11 disaster. The covert network involved in this attack is used as dataset for SpyNetMiner. The performance of SpyNetMiner is compared to a similar system called as UNICORN and other conventional algorithms. The results evidently show that SpyNetMiner outperforms all existing methodologies in covert network analysis. SpyNetMiner: An Outlier Analysis to Tag Elites in Clandestine Social Networks