Abstract

Web-Based Data Mining, Risk Forecasting and Near-Repeat Pattern Analysis
Track: Law Enforcement and Criminal Justice
Authors: Mike Urciuoli, Robert Cheetham

HunchLab software, developed with support from a National Science Foundation grant, combines ArcGIS Server technology with spatial statistics to perform crime analysis and density mapping. New features are being implemented to support risk forecasting capabilities, including near-repeat calculations. Research has demonstrated a near-repeat phenomenon in certain crime patterns for repeat victimization or retaliatory actions in or around the area where the initial parent event occurred. Near-repeat pattern analysis enables police officers to more effectively gauge both the level of increased threat and its potential duration. It has also been tested with military data to find near-repeat patterns in improvised explosive device (IED) attacks in Iraq and Afghanistan.

This presentation will focus on the development and applications of HunchLab's risk forecasting techniques using both local crime data from the Philadelphia Police Department and IED events in Afghanistan, and the optimizations needed to enable the forecasting methodologies to operate in real-time.