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.

Applications of machine-learning-based approaches in the geosciences have witnessed a substantial increase over the past few years. Here we present an approach that accounts for spatial autocorrelation by introducing spatial features to the models. In particular, we explore two types of spatial features, namely spatial lag and eigenvector spatial filtering (ESF). These features are used within the widely used random forest (RF) method, and their effect is illustrated on two public datasets of varying sizes (Meuse and California housing datasets). The least absolute shrinkage and selection operator (LASSO) is used to determine the best subset of spatial features, and nested cross-validation is used for hyper-parameter tuning and performance evaluation. We utilize Moran’s I and local indicators of spatial association (LISA) to assess how spatial autocorrelation is captured at both global and local scales. Our results show that RF models combined with either spatial lag or ESF features yield lower errors (up to 33% different) and reduce the global spatial autocorrelation of the residuals (up to 95% decrease in Moran’s I) compared to the RF model with no spatial features. The local autocorrelation patterns of the residuals are weakened as well. Compared to benchmark geographically weighted regression (GWR) models, the RF models with spatial features yielded more accurate models with similar levels of global and local autocorrelation in the prediction residuals. This study reveals the effectiveness of spatial features in capturing spatial autocorrelation and provides a generic machine-learning modelling workflow for spatial prediction.

We examined published maps containing sensitive data, and the protection methods, if any, that were used. We investigated whether the many published warnings about disclosure risk have been effective in reducing privacy risk. During an 8-year period (2005-2012), 19 journals related to GIScience, geography, spatial crime analysis, and health geography were examined. We identified 41 articles that display actual confidential information and 16 articles where confidential information is protected by the use of a geographical mask. During the investigated time frame, the numbers of articles with unmasked confidential data increased, and in total more than 68,000 home addresses were disclosed. One of the more significant findings of this study is that efforts to instill sensitivity to location privacy and disclosure risk have been relatively unsuccessful.

Reverse geocoding is defined as the extraction of textual information, such as a name or an address, from geographic coordinates. This technique is common in many geo-application scenarios, e.g., in freely available online-based mapping services. However, if personal data are mapped, confidentiality issues may arise, such as if the data are derived from a crime-incidents database. This research provides an evaluation of seven freely available Web mapping services in regard to reverse geocoding accuracy and privacy aspects. The services' textual accuracy is measured using crime data from Vienna, Austria. The accuracy associated with reverse geocoding is derived for districts with different residential densities to determine whether any error variation is associated with the underlying residential structure. Furthermore, this research investigates the reidentification risk when compromising personal information. Using a free online telephone database and addresses in a crime-incidents dataset, the probable names that are linked to crime victims are identified. The results showed that the accuracy ranges from 0% to 23% depending on the predefined precision level. Comparing seven online reverse geocoding services (TravelGIS, Google Maps, nominatim, ArcGIS, Yahoo, Bing, and Geonames), TravelGIS performs better at the highest precision level, whereas Geonames performs better at a lower precision level. However, the results showed no difference among the districts with different residential densities. As for the reidentification risk, 68% probable victims were identified from the database. This work discusses the specific tasks that users have to complete when retrieving addresses associated with crime-incident locations. The article concludes with an evaluation of the reverse geocoding process and a discussion of the privacy risk associated with it.

The General Data Protection Regulation (GDPR) protects the personal data of natural persons and at the same time allows the free movement of such data within the European Union (EU). Hailed as majestic by admirers and dismissed as protectionist by critics, the Regulation is expected to have a profound impact around the world, including in the African Union (AU). For European–African consortia conducting research that may affect the privacy of African citizens, the question is ‘how to protect personal data of data subjects while at the same time ensuring a just distribution of the benefits of a global digital ecosystem?’ We use location privacy as a point of departure, because information about an individual’s location is different from other kinds of personally identifiable information. We analyse privacy at two levels, individual and cultural. Our perspective is interdisciplinary: we draw from computer science to describe three scenarios of transformation of volunteered or observed information to inferred information about a natural person and from cultural theory to distinguish four privacy cultures emerging within the EU in the wake of GDPR. We highlight recent data protection legislation in the AU and discuss factors that may accelerate or inhibit the alignment of data protection legislation in the AU with the GDPR.

The European Union (EU) Commission’s whitepaper on Artificial Intelligence (AI) proposes shaping the emerging AI market so that it better reflects common European values. It is a master plan that builds upon the EU AI High-Level Expert Group guidelines. This article reviews the masterplan, from a culture cycle perspective, to reflect on its potential clashes with current societal, technical, and methodological constraints. We identify two main obstacles in the implementation of this plan: (i) the lack of a coherent EU vision to drive future decision-making processes at state and local levels and (ii) the lack of methods to support a sustainable diffusion of AI in our society. The lack of a coherent vision stems from not considering societal differences across the EU member states. We suggest that these differences may lead to a fractured market and an AI crisis in which different members of the EU will adopt nation-centric strategies to exploit AI, thus preventing the development of a frictionless market as envisaged by the EU. Moreover, the Commission aims at changing the AI development culture proposing a human-centred and safety-first perspective that is not supported by methodological advancements, thus taking the risks of unforeseen social and societal impacts of AI. We discuss potential societal, technical, and methodological gaps that should be filled to avoid the risks of developing AI systems at the expense of society. Our analysis results in the recommendation that the EU regulators and policymakers consider how to complement the EC programme with rules and compensatory mechanisms to avoid market fragmentation due to local and global ambitions. Moreover, regulators should go beyond the human-centred approach establishing a research agenda seeking answers to the technical and methodological open questions regarding the development and assessment of human-AI co-action aiming for a sustainable AI diffusion in the society.