Crime Generators Research Articles

Assessing Crime History as a Predictor: Exploring Hotspots of Violent and Property Crime in Malmö, Sweden

Objectives: Assessing the predictive accuracy of using prior crime, place attributes, ambient population, community structural, and social characteristics, in isolation and combined when forecasting different violent and property crimes. Method: Using multilevel negative binomial regression, crime is forecasted into the subsequent year, in 50-m grid-cells. Incidence rate ratio (IRR), Prediction Accuracy Index (PAI), and Prediction Efficacy Index (PEI*) are interpreted for all combined crime generators and community characteristics. This study is partially a test of a crude version of the Risk Terrain Modeling technique. Results: Where crime has been in the past, the risk for future crime is higher. Where characteristics conducive to crime congregate, the risk for crime is higher. Community structural characteristics and ambient population are important for some crime types. Combining variables increases the accuracy for most crime types, looking at the IRR. Taking the geographical area into account, crime history in combination with both place- and neighborhood characteristics reaches similar accuracy as crime history alone for most crime types and most hotspot cutoffs. Conclusions: Crime history, place-, and neighborhood-level attributes are all important when trying to accurately forecast crime, long-term at the micro-place. Only counting past crimes, however, still does a really good job.

Unraveling the influence of income-based ambient population heterogeneity on theft spatial patterns: insights from mobile phone big data analysis

While previous research has underscored the profound influence of the ambient population distribution on the spatial dynamics of crime, the exploration regarding the impact of heterogeneity within the ambient population, such as different income groups, on crime is still in its infancy. With the support of mobile phone big data, this study constructs an index of ambient population heterogeneity to represent the complexity of the social environment. After controlling for the effects of total ambient population, nonlocal rate, transportation accessibility, crime attractors, and crime generators, this study employs a negative binomial regression model to examine the influence of ambient population heterogeneity and different income groups on the spatial manifestations of thefts. The findings indicate that ambient population heterogeneity significantly escalates the incidence of thefts, with middle and upper-middle-income groups acting as more attractive targets, whereas the higher-income group exerts a deterrent effect. The interaction analysis shows that increased population heterogeneity contributes to social disorder, thereby amplifying the attractiveness of the ambient population to perpetrators. These conclusions highlight the crucial role of ambient population heterogeneity in explaining crime dynamics and therefore enrich the routine activity theory.

Effects of ambient population with different income levels on the spatio-temporal pattern of theft: A study based on mobile phone big data

Recent studies focus on how to accurately measure ambient population and its effects on crimes. However, limited attention has been paid to the internal heterogeneity of ambient population composition and its temporal variation. To fill these gaps, controlling for indicators of ambient population, social disorganization, crime generators and attractors, we model the effects of the share of low income, middle income, upper-middle income and high income of ambient population at the community level in the whole and specific time periods on weekdays and weekends. Results suggest that while all income groups of the ambient population contribute to theft at the community level, their contributions are not uniform. At different periods of both weekdays and weekends, there is no single income group that consistently predicts theft. Furthermore, the magnitude of the impact of each income group fluctuates with time. Our results highlight the heterogeneous segments of the ambient population, coupled with the dynamic human mobility patterns, structure the convergence of the crime triangle – motivated offenders, potential targets, and guardianship – in nuanced yet crucial ways.

Auditing the fairness of place-based crime prediction models implemented with deep learning approaches

Place-based crime prediction models implemented with deep learning leverage the spatio-temporal patterns of historical crimes, together with built-environment factors, to predict aggregate volumes of crime incidents at specific locations. Nevertheless, related work has shown that crime incident data can suffer from under-reporting bias whereby certain social groups report more crimes than others and whereby certain crimes are recorded more often by the police. In addition, under the umbrella of crime opportunity theories that suggest that human mobility can play a role in crime generation, increasing attention has been paid to the predictive power of human mobility in place-based crime models implemented with deep learning. However, prior work has shown that human mobility data collected from cell phones can suffer from sampling bias because certain population groups are not equally represented due to lack of cell phone ownership. Despite these well reported data bias concerns, no related work has yet looked into the fairness of place-based crime prediction models implemented with deep learning; nor into the effect that adding human mobility data from cell phones as a predictor might have on their fairness. In this paper, we use publicly available fine-grained crime incident and human mobility data from the US to conduct a comprehensive fairness audit across multiple cities with diverse demographic characteristics, different types of crimes and various deep learning models; and we explore potential root causes of the models' fairness metrics by looking into crime data bias, mobility data bias and algorithmic bias. Our results show that unfair predictions are pervasive across place-based crime prediction models implemented with deep-learning due to the inherent data bias in the crime incident data; that adding mobility features extracted from cell phone data decreases fairness under certain circumstances; and that part of the fairness loss can be explained by bias in the crime data, and in the predictive algorithms that might be exacerbating the data bias, with the mobility data bias having an insignificant effect on fairness loss.

Examining Train Stations as Crime Generators and the Protective Effect of “Regular” Riders

Understanding whose eyes make up the ambient population is an important development of crime opportunities research. We hypothesize that regularity at train stations will influence place-awareness and therefore the balance between potential crime offenders, targets, and victims in the ambient population. Drawing on one year of travel card and police-recorded crime data we examine the association between percentage of regular passengers and crime at 134 train stations, controlling for land use and neighborhood social demographic characteristics. We find regularity has a buffering impact on theft and property damage at stations but this effect is curvilinear in the case of theft. Regularity was not associated with assault or motor vehicle crime, we discuss implications for theory and practice.

Crime generators or social capital organizations? Examining the effects of places of worship on neighborhood crime.

Places of worship (POW) have traditionally been argued to have crime-reducing effects in neighborhoods because of their ability to produce social capital. Yet, the evidence for this proposition is surprisingly weak. Consequently, an alternative proposition, rooted in environmental criminology, suggests that POW might unintentionally operate as crime generators in neighborhoods insofar as they induce foot traffic and undermine guardianship and social control capabilities. Because of these competing propositions in combination with the limited number of studies on this topic, we conduct a block group analysis of crime, places of worship, well-established criminogenic facilities, and sociodemographic characteristics in Washington, DC. We estimate negative binomial regression models of both violent and property crime and find strong evidence for only one of the propositions, with the effects of POW being relatively strong in comparison to other predictors in the models. The implications of these findings for criminology, urban studies, and public policy are discussed.

CRIME HOTSPOT EMERGENCE IN MEXICO CITY: A COMPLEXITY SCIENCE PERSPECTIVE

We present a dynamic model based on the theories proposed by environmental criminologists to explain the emergence of crime hotspots within cities; a pervasive phenomenon that is largely independent of cities size and cultural differences. The model is defined on a multiplex network that represents a city spatial tiling with its corresponding urban transport infrastructure, allowing to explore the relation between crime hotspot locations and the network topological features. It also allows to explore the effects that cities time evolution and police checkpoints might have on the emergence of crime hotspots. For Mexico City, the model shows that heterogeneous distributions of criminal activity arise from a diffusion-driven instability, as a self-organizing process. The results obtained for this city are in line with several insights from environmental criminology, such as the relationship between urban layout and crime hotspots locations, or the conceptual label assigned to specific locations as crime generators. They also uncover new relationships between cities design and crime hotspot locations, and suggest that routine activity theory alone cannot explain the emergence of heterogeneous crime distributions.

Enhancing short-term crime prediction with human mobility flows and deep learning architectures

Place-based short-term crime prediction models leverage the spatio-temporal patterns of historical crimes to predict aggregate volumes of crime incidents at specific locations over time. Under the umbrella of the crime opportunity theory, that suggests that human mobility can play a role in crime generation, increasing attention has been paid to the predictive power of human mobility in place-based short-term crime models. Researchers have used call detail records (CDR), data from location-based services such as Foursquare or from social media to characterize human mobility; and have shown that mobility metrics, together with historical crime data, can improve short-term crime prediction accuracy. In this paper, we propose to use a publicly available fine-grained human mobility dataset from a location intelligence company to explore the effects of human mobility features on short-term crime prediction. For that purpose, we conduct a comprehensive evaluation across multiple cities with diverse demographic characteristics, different types of crimes and various deep learning models; and we show that adding human mobility flow features to historical crimes can improve the F1 scores for a variety of neural crime prediction models across cities and types of crimes, with improvements ranging from 2% to 7%. Our analysis also shows that some neural architectures can slightly improve the crime prediction performance when compared to non-neural regression models by at most 2%.

The Influence of the Environment on Officer-Involved Shootings

Research on officer-involved shootings has covered many methods and designs, with a majority focusing on macro-level approaches to understanding the shooting event. Research designs which incorporate place are limited, but what does exist demonstrates the relevance of the micro-environment to contextualize the many factors which give rise to the shooting event. This research further explores the importance of the micro-environment by examining how crime generators and attractors, conceptualized as the built environment, can be utilized to understand where police officers are more likely to discharge their weapon. Data on police shootings in Philadelphia, PA which occurred between 2015 and 2019, along with data describing the environmental landscape of the city are used in conjunction with variables previously determined to be related to shooting events. This analysis uses a three-pronged approach to understanding the relevance of the micro-environment in officer-involved shootings. Results indicate that the most important predictor of police shootings are levels of violent crime. Violent crime rates, in turn, are significantly related to the built environment. The built environment, then, has an indirect relationship to the places where police are most likely to discharge their weapon which is mediated through the violent crime rate.

Modeling the Effect of Streetscape Environment on Crime Using Street View Images and Interpretable Machine-Learning Technique.

Street crime is a common social problem that threatens the security of people's lives and property. Understanding the influencing mechanisms of street crime is an essential precondition for formulating crime prevention strategies. Widespread concern has contributed to the development of streetscape environment features as they can significantly affect the occurrence of street crime. Emerging street view images are a low-cost and highly accessible data source. On the other hand, machine-learning models such as XGBoost (eXtreme Gradient Boosting) usually have higher fitting accuracies than those of linear regression models. Therefore, they are popular for modeling the relationships between crime and related impact factors. However, due to the "black box" characteristic, researchers are unable to understand how each variable contributes to the occurrence of crime. Existing research mainly focuses on the independent impacts of streetscape environment features on street crime, but not on the interaction effects between these features and the community socioeconomic conditions and their local variations. In order to address the above limitations, this study first combines street view images, an objective detection network, and a semantic segmentation network to extract a systematic measurement of the streetscape environment. Then, controlling for socioeconomic factors, we adopted the XGBoost model to fit the relationships between streetscape environment features and street crime at the street segment level. Moreover, we used the SHAP (Shapley additive explanation) framework, a post-hoc machine-learning explainer, to explain the results of the XGBoost model. The results demonstrate that, from a global perspective, the number of people on the street, extracted from street view images, has the most significant impact on street property crime among all the street view variables. The local interpretability of the SHAP explainer demonstrates that a particular variable has different effects on street crime at different street segments. The nonlinear associations between streetscape environment features and street crime, as well as the interaction effects of different streetscape environment features are discussed. The positive effect of the number of pedestrians on street crime increases with the length of the street segment and the number of crime generators. The combination of street view images and interpretable machine-learning techniques is helpful in better accurately understanding the complex relationships between the streetscape environment and street crime. Furthermore, the readily comprehensible results can offer a reference for formulating crime prevention strategies.

The Influence, Saliency, and Consistency of Environmental Crime Predictors: A Probability Score Matching Approach to Test What Makes a Hot Spot Hot

Environmental criminological research includes several perspectives that explain how characteristics of the environment can lead to crime. This study simultaneously integrated these different perspectives, including crime generators and attractors and environmental disorder indicators, to determine the most influential, salient, and consistent predictors of micro-level crime hot spots in Indianapolis, Indiana. Hot spot and control units were compared based on the presence of spatially joined crime generators and attractors and environmental disorder indicators recorded via remote systematic social observation using Google Street View. The results indicate that crime generators and attractors tend to vary in significance, influence, saliency, and consistency across crime types. Alternatively, the presence of decay in hot spots may be the most prominent and stable environmental feature. Environmental criminology has placed an increased focus on crime generators and attractors, though, it may be time to assess the theoretical and practical importance of other criminogenic environmental elements like decay.

Identifying crime generators and spatially overlapping high‐risk areas through a nonlinear model: A comparison between three cities of the Valencian region (Spain)

The behavior and spatial distribution of crime events can be explained through the characterization of an area in terms of its demography, socioeconomy, and built environment. In particular, recent studies on the incidence of crime in a city have focused on the identification of features of the built environment (specific places or facilities) that may increase crime risk within a certain radius. However, it is hard to identify environmental characteristics that consistently explain crime occurrence across cities and crime types. This article focuses on the assessment of the effect that certain types of places have on the incidence of property crime, robbery, and vandalism in three cities of the Valencian region (Spain): Alicante, Castellon, and Valencia. A nonlinear effects model is used to identify such places and to construct a risk map over the three cities considering the three crime types under research. The results obtained suggest that there are remarkable differences across cities and crime types in terms of the types of places associated with crime outcomes. The identification of high‐risk areas allows verifying that crime is highly concentrated, and also that there is a high level of spatial overlap between the high‐risk areas corresponding to different crime types.

Agglomerative Effects of Crime Attractors and Generators on Street Robbery? An Assessment by Luojia 1-01 Satellite Nightlight

Scholars have long confirmed the agglomerative effect in retail geography. The colocation of multiple shops would encourage customers to make multiple stops in a single shopping trip. Does the agglomerative effect exist in illegal activities such as crime? What is a viable measurement of the colocation of facilities that attract or generate crime? These questions have never been explicitly addressed in crime geography. Numerous studies explain crime by the count of facilities, but the count variable ignores the size variation among the facilities. Because facilities are typically associated with lights at night, this study uses Luojia 1-01 satellite nightlight and a Gini coefficient–based adjuster to infer the agglomerative impact of crime attractors and generators in Cincinnati, Ohio. Results show that nightlights have a strong spatial correlation with the facilities and they can effectively capture large-sized and moderate-sized clusters of diverse types of crime attractors and generators. Additionally, negative binomial regression models compare the impacts of these different measures on crime by controlling potential confounding variables representing social disorganization. Results show that the nightlight models outperform the count models. Such advantage is more pronounced in areas where the crime rates are high. This is certainly an encouraging outcome, because both crime research and crime prevention tend to focus on high-crime areas. In sum, this preliminary study on the possible agglomerative effect in illegal activities reveals that nightlight data can effectively measure the agglomeration of crime attractors and generators and that the agglomerative nightlight explains crime better than the popular measure of facility counts.

Outlaw motorcycle gangs and neighbourhood crime

ABSTRACT This study examines the spatial concentration of outlaw motorcycle gang (OMCG) incidents in Melbourne, Australia. The aim of this paper is to explore how OMCG incidents cluster and to identify the neighbourhood-specific characteristics and crime attractors that are associated with the presence of such incidents. We compare the clustering of OMCG incidents with several other non-OMCG crimes to investigate whether OMCG incidents are likely to occur in areas where there is a broader history of criminogenic activity. The results indicate that disadvantage and residential instability are statistically significant for both OMCG and non-OMCG crime incidents. Ethnic diversity was only significant for non-OMCG crime incidents. Interestingly, crime attractors, specifically licensed premises and tattoo parlours, increased the prevalence of non-OMCG crime types. The study finds greater support for social disorganisation theory and the concentration of OMCG crime than for the presence of crime attractors and crime generators.

Location Choice of Snatching Offenders in Chennai City

ObjectivesWe aim to test the applicability of crime pattern theory in an Indian urbancontext by assessing the effects of offender residence, prior offending locations andpresence of crime generators and crime attractors on where offenders commit offences.MethodsThe data comprise 1573 police-recorded snatching offenses committed by 1152 identified offenders across the 201 wards of Chennai City. We used discrete crime location choice models to establish the choice criteria that snatching offenders use when they decide where to offend. Data on the locations retail businesses, religious and transportation facilities were collected using Google location services.ResultsThe results confirm that snatching offenders prefer to target locations closerto their residence and that they prefer to re-offend at or near their prior offendinglocations. The findings also demonstrate that some but not all crime attractors andgenerators influence the location choice of snatching offenders.ConclusionsBy replicating in an Indian context previously published crime locationchoice findings, our findings support the generality of crime pattern theory. We discusslimitations and make suggestions for future investigations.

Small Local versus Non-Local: Examining the Relationship between Locally Owned Small Businesses and Spatial Patterns of Crime

In the current study, we theorized that businesses in place are subject to two processes: a crime generator effect in which they heighten crime due to increased opportunities and a crime inhibition effect in which certain types of businesses can increase guardianship capability. We explicitly compare the different effects of local vs. non-local and small vs. large businesses on crime in street segments using the data in cities across the Los Angeles metropolitan region by estimating a set of negative binomial regression models for small local, large local, small non-local, and large non-local consumer facing businesses (Retail, Restaurants, Food/Drug Stores, and Services) for violent and property crime. Although we found that most of the business coefficients were positive, local businesses, and particularly small local businesses, have considerably smaller crime-enhancing effects for both violent and property crime.

The role of geographical juxtaposition in Kabalagala

ABSTRACT Geographical juxtaposition has been cited as a concept least studied by CPTED experts. Additionally, Urban planners and designers seem to have sparse knowledge on the applications of the geographical juxtaposition concept. This study evaluates the relationship between bars and pubs and the crime prevalent streets of Kabalagala township CBD. Kabalagala township as a crime hotspot within Kampala City presents appropriate conditions to support this study and generate more interest in CPTED. Kabalagala is an entertainment hotspot with numerous bars and pubs, gambling parlors, and other entertainment businesses. This paper focuses on analyzing the built environment of Kabalagala in terms of the principle of geographical juxtaposition. The outcome of polling property/business owners and police/private security firms suggested that bars and pubs created conditions necessary for crime generation, attraction, and facilitation in unison with the geographical juxtaposition concept and may be linked to property crime on adjacent streets.

Do crime generators exist in a developing context? An exploratory study in the township of Khayelitsha, South Africa

Crime generators are places that increase the risk of crime occurring due mainly to the fact that they attract a large number of motivated offenders and suitable targets to a particular location. Research identifying crime generators has however almost exclusively been confined to the United States and Europe with much less known about the ability of businesses, institutions, and/or public facilities to generate crime in less developed contexts, particularly in Africa. In this study we explore whether four different types of facilities act as crime generators in a setting axiomatically different to the West, namely Khayelitsha, a township located on the urban periphery of Cape Town in South Africa. Using intensity value analysis (IVA), we examine the intensity of assault, robbery and rape around four different types of facilities (schools, recreational hubs, transport interchanges, and alcohol outlets) relative to other random locations in Khayelitsha. The results of our research suggest that schools, and to a lesser extent transport interchanges and recreational hubs, may act as crime generators in the township. Possible explanations for these findings conclude.

Evaluation of Crime and Safety in Lagos Metropolitan Parks

The principle of Crime Prevention through Environmental Design (CPTED) was applied to examine the fear of crime on park users at selected urban parks in Lagos State, Nigeria. The objectives were to evaluate the types of security measures, as well as the types and frequency of crime in urban parks. Incidental sampling technique was employed to gather information from 300 patrons in the parks. In addition, site observations, crime mapping, and interviews with park users as well as managers were used to collect data. Descriptive statistical methods were used for data analysis. Findings revealed that the parks had very few crime generators. In addition, low levels of crime occurrences were observed due to the physical settings and their locations within the city. Although, the majority of the park users (90.4%) felt safe in the parks but still had fear of crime relating to crimes that they had experienced somewhere apart from the parks under study. The study concluded that security-related environmental design attributes had a significant effect on the users’ safety perception of the parks. Suggestions on how to improve safety conditions in the parks were made.

Forecasting Spatio-Temporal Variation in Residential Burglary with the Integrated Laplace Approximation Framework: Effects of Crime Generators, Street Networks, and Prior Crimes

ObjectivesWe investigate the spatio-temporal variation of monthly residential burglary frequencies across neighborhoods as a function of crime generators, street network features and temporally and spatially lagged burglary frequencies. In addition, we evaluate the performance of the model as a forecasting tool.MethodsWe analyze 48 months of police-recorded residential burglaries across 20 neighborhoods in Amsterdam, the Netherlands, in combination with data on the locations of urban facilities (crime generators), frequencies of other crime types, and street network data. We apply the Integrated Laplace Approximation method, a Bayesian forecasting framework that is less computationally demanding than prior frameworks.ResultsThe local number of retail stores, the number of street robberies perpetrated and the closeness of the local street network are positively related to residential burglary. Inclusion of a general spatio-temporal interaction component significantly improves forecasting performance, but inclusion of spatial proximity or temporal recency components does not.DiscussionOur findings on crime generators and street network characteristics support evidence in the literature on environmental correlates of burglary. The significance of spatio-temporal interaction indicates that residential burglary is spatio-temporally concentrated. Our finding that recency and proximity of prior burglaries do not contribute to the performance of the forecast, probably indicates that relevant spatio-temporal interaction is limited to fine-grained spatial and temporal units of analysis, such as days and street blocks.