Community-led solutions to crime and disorder hold considerable promise, but also come with challenges. Community leaders and residents are likely to have a much richer, empathetic understanding of local problems, but are also likely to lack objective data. The infrastructure to collect and use data to support crime prevention resides almost exclusively with law enforcement agencies. Community-led solutions therefore require access to public or self-generated sources of data if they are to be successful. The research team is developing a low-cost acoustic gunshot detection network for use by civilian community intervention workers in Los Angeles. The project is an example of 'data democratization.' Communities take the lead in data collection and dissemination as part of solving community problems.

This research network uses off-the-shelf hardware and custom deep learning algorithms to recognize gunshots in real-time, including where and when they occur. Community intervention workers control the placement of detectors in the community and use a smartphone-based alert system, developed by the research team, to support their violence prevention efforts. Current evidence suggests that community intervention workers can stop violent retaliations if they are able to respond quickly when violence occurs. The research team uses controlled experiments and causal inference methods to assess the impact of the gunshot detection network on community-led violence prevention. The project builds trust in data and an appreciation for the inclusion of experimental methods in community-led models. Demonstrating successful data-driven solutions is essential to the development of smart and connected communities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.