Centro de Documentação da PJ
Monografia
 CD258 |
Predictive policing [Documento electrónico] : the role of crime forecasting in law enforcement operations / Walter L. Perry ..[et al.].- Santa Monica : RAND Corporation, 2013.- 1 CD-ROM ; 12 cm. - (Safety and Justice Program)
The research described in this report was sponsored by the National Institute of Justice and conducted in the Safety and Justice Program within RAND Justice, Infrastructure, and Environment.
ISBN 978-0-8330-8148-3
GEOGRAFIA CRIMINAL, MAPEAMENTO DO CRIME, ANÁLISE CRIMINAL, SISTEMA DE INFORMAÇÃO GEOGRÁFICA, INFORMAÇÃO COMPUTORIZADA, ESTADOS UNIDOS
Preface. Figures. Tables. Summary. Acknowledgments. Abbreviations. Chapt er One. Introduction. What is Predictive Policing?. A criminological justification for predictive policing: why crime is “predictable”. A brief history of predictive policing. Background.. Training. Study objectives and methods. Objectives. Approach. The nature of predictive policing: this is not minority report. A taxonomy of predictive methods. Prediction-led policing processes and practices. Data collection. Analysis. Police operations. Criminal response. About this report. Chapter Two. Making predictions about potential crimes. Notes on software. Hot spot analysis and crime mapping. Grid mapping. Covering ellipses. Single and dual Kernel Density Estimation. Heuristic methods. Regression methods. Types of relationships. Selecting input variables. Leading indicators in regression (and other) models. A regression example. Data mining (predictive analytics). Clustering. Classification. Training and testing a model. Near-repeat methods. Spatiotemporal analysis. Basics of spatiotemporal analysis. Heat maps. Spatiotemporal modeling using the generalized additive model. Seasonality. Risk terrain analysis. A heuristic approach: risk terrain modeling. A statistical approach to risk terrain analysis. Discussion of risk terrain analysis approaches. Prediction methods. Chapter Three. Using predictions to support police operations. Evidence-based policing. Taking action on hot spots in Washington, D.C. Koper curve application in Sacramento. Investigating convenience store robberies in Chula Vista, California. Predictive policing in context: case studies. Shreveport, Louisiana: Predictive intelligence–led operational targeting. Memphis, Tennessee: crime reduction utilizing statistical history. Nashville, Tennessee: integrating crime and traffic crash data. Baltimore, Maryland: crash-crime project. Iraq: locating IED emplacement locations. Minneapolis, Minnesota: micro crime hot spots. Charlotte-Mecklenburg County, North Carolina: foreclosures and crime. Crime maps: community relations. Police actions. Chapter Four. Using predictions to support investigations of potential offenders. Protecting Privacy rights and civil liberties. Predictive policing symposium assessment. Privacy under the Fourth Amendment of the U.S. Constitution. Privacy with respect to policing intelligence information systems. privacy resources for the Law Enforcement Community. Dealing with noisy and conflicting data: data fusion. Heuristic and simple-model methods. More sophisticated fusion methods. Risk assessment for individual criminal behavior. Commonly used behavioral instruments. Limitations of behavioral instruments. Quebec, Canada: Assessing criminogenic risks of gang members. Pittsburgh, Pennsylvania: Predicting violence and homicide among young men. Risk assessment for organized crime behavior. Risk assessment instruments for domestic violence. Risk assessment instruments for mental health. Predictive methods: finding suspects. Basic queries. Criminal intelligence in social network analysis format. Links to Department of Motor Vehicle Registries, Pawn Data, and other registries. Anchor point analysis, or geographic profiling. Modus Operandi similarity analysis. Exploitation of sensor data. Putting the clues together: multisystem and network queries. Taking action on predictions. Identifying high-risk individuals. Identifying the most likely suspects. Prediction-based offender intervention in context. Florida State Department of Juvenile Justice: preemptive efforts. Predicting predator hunting patterns. Chapter Five. Findings for practitioners, developers, and policymakers. Predictive policing myths. Myth 1: The computer actually knows the future. Myth 2: The computer will do everything for you. Myth 3: you need a high-powered (and expensive) model. Myth 4: Accurate predictions automatically lead to major crime reductions. Predictive policing pitfalls. Pitfall 1: Focusing on prediction accuracy instead of tactical utility. Pitfall 2: Relying on poor-quality data. Pitfall 3: Misunderstanding the factors behind the prediction. Pitfall 4: Underemphasizing assessment and evaluation. Pitfall 5: Overlooking civil and privacy rights. A Buyer's Guide to Predictive Policing. A Developer’s Guide to Predictive Policing. Desired capabilities. Controlling the hype. Access and affordability. A Crime Fighter’s Guide to using Predictive Policing: emerging practices. Supporting the establishment of predictive policing processes. Practices for data collection and analysis: creating shared situational awareness. Operations: Key sources of information on interventions. Promising general practices for operations.. Conclusions. About the authors. Bibliography.