Search

Bij haar werkzaamheden maakt de politie gebruikt van voorspellende algoritmes. In de literatuur wordt de vrees uitgesproken dat het gebruik van voorspellende algoritmes sneller leidt tot het inzetten van (vrijheidsbenemende) dwangmiddelen tegen burgers. Deze inzet is alleen gerechtvaardigd als er sprake is van een redelijke verdenking. Welke bijdrage kan een voorspelling van een algoritme leveren aan het ontstaan van een redelijke verdenking? Voor het beantwoorden van deze vraag wordt een vergelijking gemaakt tussen de voorspelling van een algoritme en andere feiten en omstandigheden waar een redelijke verdenking op kan worden gebaseerd, zoals algemene ervaringsregels en tips uit anonieme bronnen.

This chapter covers geographic profiling, from its historical roots to the current theoretical propositions and practical applications. From the development and theoretical roots, such as theories of spatial behavior and main protagonists, Norris discusses whether geographic profiling is a method in its own right, or whether it forms part of an overall investigative strategy.

Predictive policing kan een duidelijke meerwaarde hebben als ondersteuning van het politieproces, maar dat is niet altijd het geval. Bewustzijn van de beperkingen van algoritmes is daarom belangrijk, stellen Reinder Doeleman, René Melchers, Lauren Waardenburg en Dick Willems.

Predictive policing has become a new panacea for crime prevention. However, we still know too little about the performance of computational methods in the context of predictive policing. The paper provides a detailed analysis of existing approaches to algorithmic crime forecasting. First, it is explained how predictive policing makes use of predictive models to generate crime forecasts. Afterwards, three epistemologies of predictive policing are distinguished: mathematical social science, social physics and machine learning. Finally, it is shown that these epistemologies have significant implications for the constitution of predictive knowledge in terms of its genesis, scope, intelligibility and accessibility. It is the different ways future crimes are rendered knowledgeable in order to act upon them that reaffirm or reconfigure the status of criminological knowledge within the criminal justice system, direct the attention of law enforcement agencies to particular types of crimes and criminals and blank out others, satisfy the claim for the meaningfulness of predictions or break with it and allow professionals to understand the algorithmic systems they shall rely on or turn them into a black box. By distinguishing epistemologies and analysing their implications, this analysis provides insight into the techno-scientific foundations of predictive policing and enables us to critically engage with the socio-technical practices of algorithmic crime forecasting.

The research report looks at place-oriented predictive policing technologies used in France. After describing the goal of this report and our methodology, we offer some background on the institutional structure of French policing. We then turn to case studies, looking in detail at three place-oriented predictive policing technologies used in the country: PAVED, a tool developed by the French Gendarmerie; Smart Police, a product sold by a French startup by the name of Edicia to local police forces; M-Pulse, a project once labelled as a "Big Data Observatory for Public Tranquility'' developed by the city of Marseille in partnership with the company Engie Solutions. Our conclusion is that while these systems apparently fail to meet their alleged objectives – with no to negligible impact on crime –, they come with significant risk of leading law enforcement agencies to abuse their powers by overstepping legal boundaries and over-policing marginalized populations, especially in a context marked by a lack of evaluation, faulty oversight and elusive transparency., Ce rapport de recherche s'intéresse aux technologies de police prédictive fondées sur une approche géographique (« hotspots »). Après avoir décrit l'objectif de cette recherche et notre méthode d'enquête, nous procédons à quelques rappels sur la structure institutionnelle de la police française. Les études de cas examinent ensuite trois technologies de police prédictive : PAVED, un outil développé par la Gendarmerie nationale ; Smart Police, un produit vendu par la startup Edicia aux forces de police municipale ; M-Pulse, un projet anciennement baptisé « Observatoire Big Data pour la Tranquillité Publique » et développé par la mairie de Marseille en partenariat avec la société Engie Solutions. En conclusion, nous soulignons que, alors même que ces systèmes ne parviennent apparemment pas à atteindre leurs objectifs – avec un impact nul ou négligeable sur la criminalité –, ils présentent un risque important de conduire les forces de l'ordre à abuser de leurs prérogatives en outrepassant les limites légales et en accentuant la surveillance de populations en proie à des discriminations structurelles, a fortiori dans un contexte marqué par l'absence d'évaluation, la défaillance des organismes de contrôle et un manque de transparence.

Big data and algorithms are the glue of the digital surveillance capitalism. Software runs in several social domains, which have an impact on our lives: from health care to employment and university enrolment, to bank loans and insurance policy conditions. What makes big data analytics especially appealing for the crime control agencies is the promise of detailed algorithmic prediction and pre-emption. The chapter shows the shift in the knowledge–power nexus in policing and criminal justice due to the big data revolution. The idea of predictive policing is “to issue crime forecasts in the same way as the Weather Service issues storm alerts” and to disrupt the “production cycle” of crime. The idea of automated justice is to vaporise biases, heuristics and to confine fundamentally value-based decisions to “clean and pure” mathematical reason. The chapter begins with the fascination with numbers, which started at the “birth” of criminology as a science in the nineteenth century and presents contemporary critiques of “traditional” nation-oriented crime surveys. It continues by showing how big data attaches to penal power and how it automates governance. In the central part, the chapter focuses on automated policing programs and automatisation in criminal justice systems by presenting several programs that are already being used or tested, for example, for the prevention of payment card fraud by means of skimming; for the prediction of crime with predictive software; the use of algorithms to predict the recidivism of parolees. It concludes by presenting several risks of the transition to automated justice and policing as regards encroachment on fundamental liberties and the democratic division of power.

Background Predictive policing and crime analytics with a spatiotemporal focus get increasing attention among a variety of scientific communities and are already being implemented as effective policing tools. The goal of this paper is to provide an overview and evaluation of the state of the art in spatial crime forecasting focusing on study design and technical aspects. Methods We follow the PRISMA guidelines for reporting this systematic literature review and we analyse 32 papers from 2000 to 2018 that were selected from 786 papers that entered the screening phase and a total of 193 papers that went through the eligibility phase. The eligibility phase included several criteria that were grouped into: (a) the publication type, (b) relevance to research scope, and (c) study characteristics. Results The most predominant type of forecasting inference is the hotspots (i.e. binary classification) method. Traditional machine learning methods were mostly used, but also kernel density estimation based approaches, and less frequently point process and deep learning approaches. The top measures of evaluation performance are the Prediction Accuracy, followed by the Prediction Accuracy Index, and the F1-Score. Finally, the most common validation approach was the train-test split while other approaches include the cross-validation, the leave one out, and the rolling horizon. Limitations Current studies often lack a clear reporting of study experiments, feature engineering procedures, and are using inconsistent terminology to address similar problems. Conclusions There is a remarkable growth in spatial crime forecasting studies as a result of interdisciplinary technical work done by scholars of various backgrounds. These studies address the societal need to understand and combat crime as well as the law enforcement interest in almost real-time prediction. Implications Although we identified several opportunities and strengths there are also some weaknesses and threats for which we provide suggestions. Future studies should not neglect the juxtaposition of (existing) algorithms, of which the number is constantly increasing (we enlisted 66). To allow comparison and reproducibility of studies we outline the need for a protocol or standardization of spatial forecasting approaches and suggest the reporting of a study’s key data items.