Interactive analysis of computer crimes - Computer

P olice departments and other law enforcement agencies continually analyze vast amounts of criminal incident data to better understand crime in their jurisdictions, to identify significant changes in crime levels, to plan community and neighborhood responses to crime, to investigate incidents, and to apprehend perpetrators. As the number of crimes increases and as resources—particularly human resources—for crime analysis remain relatively constant, law enforcement agencies are increasingly turning to automated tools to examine multiple incident reports and analyze criminal activities in their areas based on geographical location, time, and type. The increase in computer crimes over the past decade requires enhancing the sophistication of crime analysis tools to address new types of crimes. According to Douglas Gansler, a state’s attorney in Montgomery County, Maryland, the range and extent of computer crimes is staggering, including “cyberstalking, investment fraud, child sexual exploitation, information theft, domestic and international terrorism, copyright piracy, system deception, hate crimes, elder abuse, and plain old fraud.”1 As the Internet grows in popularity and as e-commerce becomes a projected multi-trillion-dollar component of the economy, the number of computer crimes will rise accordingly. While regional crime analysis systems for conventional crimes are just now appearing, the Internet requires at least national and, ideally, global analysis tools. Further, “Internet time” requires crime analysis at faster rates and in significantly smaller time intervals than the days, weeks, and months typical of conventional crime analysis, and “Internet location” requires tools that are not fixed on geographical location. Our computer crime analysis system builds directly on systems we researched and built for conventional crime analysis over large regions. These systems can link criminal activities by location, time, and method; they also can detect significant changes in criminal activity and discover criminal preferences to aid in predicting future threats.