While existing crime prediction and prevention methods focus on the location of the crimes to detect “hot-zones”, this project takes a fundamentally different, data-driven approach towards integrated multi-scale data analytics for identifying the characteristics and features of crime-prone environments. Unlike state-of-the-art crime forecasting methods which mainly rely on historical crime activity data, the proposed project will develop scalable data-driven algorithms that will adapt to the constantly evolving state of criminal activity by continuously learning from a rich set of spatial and demographic features, including traffic, spatial attributes, socio-economic characteristics of neighborhoods, and current time, as well as context. uch a rich set of features will enable fine-grained understanding of criminal activity and correlation between crime types.
The output of the proposed data-driven models will feed a novel multi-objective optimization formulation that will be used for the integrated optimization of personnel positioning, patrol scheduling and safest route calculation. The proposed project brings together a unique team of experts in data-driven modeling, machine learning, criminology, and social and behavioral science. The resulting decision support environment, will be transferred to the USC Department of Public Safety (DPS), the Los Angeles Police Department (LAPD), and South Park Business Improvement District (SPBID) for integration with their systems to enable decision makers to choose the best course of action at any given time.
The code repositories for this project can be found here.