Publications

Knowledge Discovery for Counterterrorism and Law Enforcement
David Skillicorn, Queen's University
Chapman and Hall, 2008.

Most of the research aimed at counterterrorism, fraud detection, or other forensic applications assumes that this is a specialized application domain for mainstream knowledge discovery. Knowledge discovery changes completely when the datasets being used have been manipulated in order to conceal some underlying activity. The book operates from the premise that detection algorithms must be rethought to be effective in this domain, and presents a new approach based on cutting-edge analysis for use in adversarial settings.

The book focuses on four main forms of knowledge discovery: prediction, clustering, relationship discovery, and textual analysis. It discusses opportunities for concealment that are available to criminals and reveals some of the tactics that can aid in detecting them. The book also supplies a preview of technologies currently under development and describes how they will fit in to existing approaches to knowledge discovery.