There are spatial and temporal features to crimes, and today's data analysis tools allow for the display and examination of this digital data. Increasingly, police departments are beginning to use new mapping applications and geographic information systems (GIS) to measure and analyze the spatial relationships of this criminal data. With these new graphical tools, analysts are able to examine place-based crime and develop contiguity matrices to view relationships between criminals and victims. In addition to these analytical advances, computerized police records management systems and computer-aided dispatch (CAD) systems of citizen calls to police make it possible to systematically quantify varying levels of criminal activity and types at different locations within a city.

There are two methods by which crimes can be mapped. One is human-driven, where an analyst interactively explores the features of crimes by different dimensions, such as type of crime by time of day or day of week. This is a top-down approach to mapping criminal activity. The second method of mapping crime, the machine-driven method, where criminal data is autonomously clustered using a data mining tool, such as a SOM neural network. This type of map is driven by the criminal data itself and may lead to the discovery of new, previously unseen patterns and can provide very important insights that a human-driven analysis might miss. This is a bottom-up approach to mapping criminal activity. In the end both the top-down and bottom-up approaches should be considered for they can compliment each other; the bottom-up approach may...