Patricia L. Brantingham

Nodes, paths and edges: Considerations on the complexity of crime and the physical environment

Crime has long been thought to be intimately associated with the physical environment in which it occurs. Theoretical and empirical developments over the past 20 years demonstrate that this relationship is complex and varies substantially at different levels of spatial and temporal resolution. Research on the distribution of property crimes in time and space resonates with research on the target selection processes of offenders to suggest that crime is strongly related to aggregate elements of the perceived physical environment: nodes, paths, edges and an environmental backcloth. The relationship between crime and the physical environment is mediated through individual awareness and action spaces. This implies a series of research issues and crime control policies for future exploration.

Mobility, Notoriety, and Crime: A Study in the Crime Patterns of Urban Nodal Points

Crime has long been known to be spatially patterned at many levels of aggregation. Contemporary explanations of this fact within urban areas assume that it is the result of interactions between the physical distribution of opportunities for crime, transportation flow patterns, and the awareness spaces of potential criminals. Data from a small city in Western Canada are used to conduct a simple test of the plausibility of this theoretical assumption for the crime of commercial burglary. The assumption is generally supported by the data.

A computational model for simulating spatial aspects of crime in urban environments

In this paper, we present a novel approach to computational modeling of social systems. By combining the abstract state machine (ASM) formalism with the multi-agent modeling paradigm, we obtain a formal semantic framework for modeling and integration of established theories of crime analysis and prediction. We focus here on spatial and temporal aspects of crime in urban areas. Our work contributes to a new multidisciplinary research effort broadly classified as Computational Criminology.

Crime Analysis at Multiple Scales of Aggregation: A Topological Approach

Patterns in crime vary quite substantially at different scales of aggregation, in part because data tend to be organized around standardized, artificially defined units of measurement such as the census tract, the city boundary, or larger administrative or political boundaries. The boundaries that separate units of data often obscure the detailed spatial patterns and muddy analysis. These aggregation units have an historic place in crime analysis, but increasing computational power now makes it possible to start with very small units of analysis and to build larger units based on theoretically defined parameters. This chapter argues for a crime analysis that begins with a small spatial unit, in this case individual parcels of land, and builds larger units that reflect natural neighborhoods. Data are limited in these small units at this point in time, but the value of starting with very small units is substantial. An algorithm based on analysis of land unit to unit similarity using fuzzy topology is presented. British Columbia (BC) data are utilized to demonstrate how crime patterns follow the fuzzy edges of certain neighborhoods, diffuse into permeable neighborhoods, and concentrate at selected high activity nodes and along some major streets. Crime patterns that concentrate on major streets, at major shopping centers and along the edges of neighborhoods would be obscured, at best, and perhaps missed altogether if analysis began with larger spatial units such as census tracts or politically defined neighborhood areas.

Modeling Criminal Activity in Urban Landscapes

Computational and mathematical methods arguably have an enormous potential for serving practical needs in crime analysis and prevention by offering novel tools for crime investigations and experimental platforms for evidence-based policy making. We present a comprehensive formal framework and tool support for mathematical and computational modeling of criminal behavior to facilitate systematic experimental studies of a wide range of criminal activities in urban environments. The focus is on spatial and temporal aspects of different forms of crime, including opportunistic and serial violent crimes. However, the proposed framework also provides a basis to push beyond conventional empirical research and engage the use of computational thinking and social simulations in the analysis of terrorism and counter-terrorism.

Co-offending Network Mining

We propose here a computational framework for co-offending network mining defined in terms of a process that combines formal data modeling with data mining of large crime and terrorism data sets as gathered and maintained by law enforcement and intelligence agencies. Our crime data analysis aims at exploring relevant properties of criminal networks in arrest-data and is based on 5 years of real-world crime data that was made available for research purposes. This data was retrieved from a large database system with several million data records keeping information for the regions of the Province of British Columbia. Beyond application of innovative data mining techniques for the analysis of the crime data set, we also provide a comprehensive data model applicable to any such data set and link the data model to the analysis techniques. We contend that central aspects considered in the work presented here carry over to a wide range of large data sets studied in intelligence and security informatics to better serve law enforcement and intelligence agencies.

Burglar Mobility and Crime Prevention Planning

Burglary is a relatively frequent crime in North America with serious finan cial and personal/psychological consequences. The rate of reported burglar ies in the United States was 1,632.1 per 100,000 population in 1981. In Canada the rate was 1,518.2 per 100,000 population in the same year. About two-thirds of reported burglaries were residential break-ins in each country in 1981. Victimization surveys indicate that the incidence of resi dential burglary is even higher. The estimate provided in the United States National Crime Survey is 87.9 per 1,000 households in 1981.

The social impact in a high-risk community: A cellular automata model

a b s t r a c t This research examines the spread of criminal behavior and hard drug consumption using a mathemat- ical approach called cellular automata (CA). This CA model is based on two behavioral concepts. Firstly, peer association impacts criminal involvement. Secondly, addiction can heighten criminal activity. The model incorporates four types of actors who interact in a high-risk social community and one interven- tion method. The actors exert a social influence on each other by encouraging or discouraging drug use and criminal behavior. The intervention method called Incapacitation has a probabilistic impact on the individuals in the model. The results identify the threshold where positive influences on a population reduce the number of high-rate offenders in the community. These results are discussed to further the knowledge about the social influences in a high-risk community and how these influences can effect decisions on offender management.

Uncovering the Spatial Patterning of Crimes A Criminal Movement Model (CriMM)

Objectives: The main objective of this study was to see if the characteristics of offenders’ crimes exhibit spatial patterning in crime neutral areas by examining the relationship between simulated travel routes of offenders along the physical road network and the actual locations of their crimes in the same geographic space. Method: This study introduced a Criminal Movement model (CriMM) that simulates travel patterns of known offenders. Using offenders’ home locations, locations of major attractors (e.g., shopping centers), and variations of Dijkstra’s shortest path algorithm we modeled the routes that offenders are likely to take when traveling from their home to an attractor. We then compare the locations of offenders’ crimes to these paths and analyze their proximity characteristics. This process was carried out using data on 7,807 property offenders from five municipalities in the Greater Vancouver Regional District (GVRD) in British Columbia, Canada. Results: The results show that a great proportion of crimes tend to be located geographically proximal to the simulated travel paths with a distance decay pattern characterizing the distribution of distance measures. Conclusion: These results lend support to Crime Pattern Theory and the idea that there is an underlying pattern to crimes in crime neutral areas.

Sentencing Disparity: An Analysis of Judicial Consistency

This paper reports the results of an analysis of judicial disparity in the sentencing of persons represented by legal-aid lawyers. Because the socioeconomic characteristics of legal-aid clients are fairly uniform, the analysis of such cases made it possible to explore the influence of case facts, system factors, and the judicial disparity of the sentences given in relatively similar situations. The analysis finds that case facts and offender characteristics, particularly prior record, are good predictors of sentence type and excellent predictors of sentence length. While there was some indication of judicial inconsistency in sentence-type decisions, that is, unexplained variation from case to case, there was little indication of strong individual judicial bias across the cases used in the analysis.