Applying Artificial Intelligence to Person-Based Policing Practices

The applicant proposes to develop an artificial intelligence (AI) tool (using agent-based modeling methods) to improve the effectiveness of the person-based focused deterrence strategy employed by the Hampton (VA) Police Division (HPD) and test the tool using an RCT (n = 240 high-risk individuals) for a 12-month period. The tool, based on crime intelligence and HPD datasets, will be used to predict risk behaviors in different environments. The tool will be evaluated on nine outcomes comparing control (using HPDs current method for identification and providing information to officers about high-risk individuals) and treatment groups (that will be further analyzed using the AI tool to provide additional information about scenarios with real-time predictions): (1) is an identified high-risk individual identified as a suspect; (2) is a high-risk individual later determined to be non-high risk; (3) number of citations per individual; (4) number of arrests; (5) number of victims per individual; (6) amount of time spent per case; (7) number of cases cleared per individual; (8) number of connections or associates identified per individual; and, (9) type and amount of evidence with probative value per individual. Once completed the tool will be available for other departments to train and implement using their jurisdictions data. Note: This project contains a research and/or development component, as defined in applicable law," and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14). CA/NCF