Applying data analytics to comprehensive linked police records

The West Midlands Police have recently developed a system to draw together over 40 data sets in real-time, including information on convictions, sentences, police custody, intelligence, 'stop and search', and forensic systems. In parallel, a 'static' dataset combines 13 such sources, matching and combining numerous data points into more than 5 million individual person records. Together these systems provide a unique opportunity to undertake retrospective longitudinal analysis and then apply emergent models prospectively. This provides potential for research that can be immediately impactful, being designed and undertaken to solve current and emerging policing priorities. The data is amenable to a range of analytic techniques. For example, machine learning techniques have been applied, creating artificial neural networks, so as to model the complex nonlinear relationships between interconnected inputs and outputs. Interpretation of the models enables the identification of modifiable factors that might be targeted by police or other agencies, though also requires critical consideration to the social construction of, and therefore biases inherent in police data. The data has a myriad of potential applications; ideas include, but are not restricted to, violence prevention, intergenerational criminalisation and victimisation, and evaluation of impact of specific interventions or events.