Striking back at Gun Related Crime: Creating a Novel Approach to Gun Related Crime through Social Network Analysis.

With gun related crime rates increasing in Tucson, Arizona over the past few years, the Tucson Police Department (TPD) is committed to the advancement of technological interventions to combat this growing problem. One-third of gun-related crime (GRC) cases within TPD go unassigned due to a lack of sufficient investigative information, and this number is increasing. The need to solve cases more effectively is pressing in the context of equity and equal access to justice as violence in Tucson disproportionately impacts communities of color. TPD aims to adopt innovative technology to tackle the escalating rates of GRC and enhance the effectiveness of investigations. TPD is seeking to use cutting-edge, non-relational databases to connect people, places, and societal problems across latent networks and strategically deploy investigative resources to disrupt the most at-risk connections. TPD's objective is to enhance its social network analysis and natural language processing capability by integrating both vector and graph databases, thereby identifying case-relevant information that would have otherwise gone unnoticed.

The project will span three years and have three main goals. First, it will test the utility of new technologies to connect unstructured and disparate criminal intelligence data, support social network analysis, and design machine learning solutions to support GRC investigations. Second, the proposal will help to provide clearer evidence of the connection between technology derived intelligence and the disruption of criminal networks. Third, the proposal would advance the state of policing practice, accountability, and science by developing protocols and thresholds for applying the new technologies to people and places. This will be achieved through the academic partnership with the Collaborative Solutions Lab, a research enterprise headed by social scientists from the University of Arizona, for the development of best practices and evaluations.

The projected results of the project involve a rise in effective GRC clearance rates, increased efficiency in investigative resource allocation, and a better understanding of the equitability of  machine learning technology. The project will serve the community of Tucson and particularly the communities of color that disproportionately suffer from violence. The intended beneficiaries of the project are the Tucson community and law enforcement agencies across the United States, which can use the protocols and these thresholds developed in the project to improve their policing practices.