Identifying key players in a network of child exploitation websites using Principal Component Analysis

Abstract

One of the main objectives of this study is to help prioritize targets for law enforcement by analyzing online websites hosting child exploitation material and finding key players within. Key players are defined as websites that display a combination of high connectivity and a lot of hardcore material and would provide the most disruption in a network if they were to be removed. In this study, various strategies based on Principal Component Analysis are presented to identify those nodes that act as the key players in an online child exploitation network. For evaluating the results of these strategies, we consider the results of various attack strategies. The measures for evaluation are the density, clustering coefficient, average path length, diameter, and the number of connected components in the resulting network. The results show that the strategies proposed are more successful at reducing all of the outcome measures than existing strategies.

Keywords

• social network analysis
• child exploitation
• network disruption
• attack strategy
• web content