Forecasting Crimes Using Autoregressive Models

摘要: As a result of steadily increasing urbanization, by 2030 more than sixty percent the global population will live in cities. This phenomenon is stimulating significant economic and social transformations, both positive (such as, increased opportunities offered urban areas) negative crime pressures on city budgets). Nevertheless, new technologies are enabling police departments to access growing volumes crime-related data that can be analyzed understand patterns trends. Such knowledge useful anticipate criminal activity thus optimize public safety resource allocation (officers, patrol routes, etc.) through mathematical techniques predict crimes. paper presents an approach, based auto-regressive models, for reliably forecasting trends areas. In particular, main goal work design predictive model forecast number crimes happen rolling time horizons. case study, we present analysis performed area Chicago, using variety open sources available exploration examination University Chicagos Plenario platform. Experimental evaluation shows proposed methodology predicts with accuracy 84% one-year-ahead forecasts 80% two-year-ahead forecasts.