Computational Criminology

Crime and terrorism in the 21^st century call for advancement in the modeling and simulation of criminal events in the complex environment. This presentation reviews the field of computational criminology, an emerging blend of criminology, computer science and applied mathematics. Modern concerns about public safety and security include a focus on a range of events from less serious everyday crimes like shoplifting through to personal violent crimes like homicide and ultimately to terrorism. Underlying all of these events is a decision process or chain of steps in target identification, steps that focus first on rough and vague decisions and move towards the precise. Minor and major crimes involve people moving about in a known space in identifiable patterns to find weaknesses. The field of computational criminology involves using computational power to identify: (1) patterns and emerging patterns; (2) crime generators and crime attractors; (3) terrorist, organized crime and gang social and spatial networks as well as co-offending networks; and, (4) cybercrime. Algorithms are developed using computational topology, hyper-graphs, SNA, KDD, agent based simulations, dynamic information systems analysis and more. This presentation is designed to provide information about crime pattern theory, pattern identification and research in computational criminology. It is designed to identify research areas of potential interest to participants at the conference. Computational criminology is an emerging field that is opening doors for new and innovative approaches. The presentation will show how people (offenders and non-offenders) move about in space with a routine time and location chronologies (in physical and internet space). Anchor points develop; primary routes emerge. Navigation and rules for navigation shape both commuting patterns; shopping patterns; web sites-forums, blogs, and shared information; and crime and terrorism patterns. Crime and terrorism are not - - random; they appear to follow rules similar to those in many types of non-criminal behavior. Better understanding these rules and developing appropriate algorithms for identifying risky areas is the continuing focus of computational criminology.