Figure - uploaded by Anton Borg
Content may be subject to copyright.
Prediction performance for the different cities.

Prediction performance for the different cities.

Source publication
Article
Full-text available
The evidence that burglaries cluster spatio-temporally is strong. However, research is unclear on whether clustered burglaries (repeats/near-repeats) should be treated as qualitatively different crimes compared to spatio-temporally unrelated burglaries (non-repeats). This study, therefore, investigated if there were differences in modus operandi-si...

Contexts in source publication

Context 1
... experiments were conducted on a per city basis, as initial experiments over the complete geographical area proved unsuccessful (mean AUC around 0.55), being little better than chance. The results of the experiments presented in Table 3 had an average F 1 -score of 0.81 over all cities. Looking at individual cities, the F 1 -score ranged from 0.66 (Harrenhal) to 0.90 (Sunspear). ...
Context 2
... well as Sunspear for the F 1 -score). Overall, the box-plots (as well as the scores in Table 3) indicate that the models are capable of differentiating between near-repeat crimes and non-repeat crimes. ...
Context 3
... experiments were conducted on a per city basis, as initial experiments over the complete geographical area proved unsuccessful (mean AUC around 0.55), being little better than chance. The results of the experiments presented in Table 3 had an average F 1 -score of 0.81 over all cities. Looking at individual cities, the F 1 -score ranged from 0.66 (Harrenhal) to 0.90 (Sunspear). ...
Context 4
... well as Sunspear for the F 1 -score). Overall, the box-plots (as well as the scores in Table 3) indicate that the models are capable of differentiating between near-repeat crimes and non-repeat crimes. ...

Similar publications

Preprint
Full-text available
Vehicle hijacking is one of the leading crimes in many cities. For instance, in South Africa, drivers must constantly remain vigilant on the road in order to ensure that they do not become hijacking victims. This work is aimed at developing a map depicting hijacking spots in a city by using Twitter data. Tweets, which include the keyword "hijacking...

Citations

... In assessing and rationalising student burglary knowledge, the current study presents burglar behaviour as somewhat fixed and patterned, allowing the study to draw comparisons to student behaviour in the virtual environment. Although the preferences of burglars for end of terrace houses, small easy to conceal goods, and other behaviours discussed in this study have been demonstrated in previous research (Logie et al. 1992;Nee et al. 2015Nee et al. , 2019, it is important to acknowledge that burglary is a multi-faceted crime, with heterogeneity across offences (Borg and Svensson 2022). This may lead to behavioural differences not discussed in this study, such as a burglar with transport stealing larger high-value items. ...
Article
Full-text available
Research has highlighted gaps in the publics ‘awareness of burglar behaviour and their perceptions of vulnerability, perhaps indicating a need for further prevention training. No such research exists examining students’ awareness, but is vital given students’ relatively high levels of burglary victimisation. To provide this research, students were asked to complete a mock burglary in a simulated environment, with burglary awareness inferred based on their performance relative to previous experienced burglars. Further context was added by qualitatively analysing participants’ verbalisations during the task. Performance was also compared between first and third year undergraduates to further examine differences in awareness across student populations. No significant differences were found between the behaviour of first and third years. Nevertheless, both groups showed an overall limited understanding of burglary behaviour through their performance and verbalisations. Results demonstrate that raising awareness of victimisation and burglary behaviour is integral to improve the safety of University students.
... Targets closer to offenders' residences get offended more than distant targets due to distance decay behaviour in offending (Laukkanen 2007). This may lead to recurrent offending of burglary (Borg and Svensson 2022) by offenders living nearby the centre of the city and the resultant 'X' hotspot of burglary during night hours (Van Sleeuwen et al. 2021). This phenomenon can be attributed to the convenience and opportunity factors that come into play when offenders have easy access to potential targets in their immediate vicinity. ...
... It is plausible that offenders residing in this area strategically commit more burglaries nearby for easier planning and execution rather than remaining distant areas of the city. This, in turn, leads to near-repeat crimes/victimization (Borg and Svensson 2022) and offending nearby during night hours to minimize the risk of being caught (Van Sleeuwen et al. 2021) and also to avoid long residence-to-crime distances. This may lead to the formation of 'W' burglary hotspot even in low socioeconomic area where there is scarcity of lucrative targets. ...
Article
Full-text available
Spatio-temporal aspect is the most useful aspect of crime prevention as it is patterned and predictable. This study is mainly conducted to explore the spatio-temporal hotspots of burglary and to test whether spaces of burglary are constrained by time in Ahmednagar city. The temporal hotspots are identified by their frequency in hours. Spatial hotspots within identified temporal hotspots are visualized through GIS maps prepared using Kernel density estimation with natural break methods. The current results and their probable causation are compared with the theories in crime geography and discussed thoroughly. The burglary hotspots in Ahmednagar city are found to vary by space–time together than individual space or time. Spaces of burglary hotspots in Ahmednagar city are found unstable over time indicating that spatial hotspots of burglary are constrained by temporal hotspots. Consistent with existing literature, probable causation behind spatio-temporal variation and temporal constraint of burglary hotspots is closely associated with the combined effect of the land use pattern of the area, movement of mass population within the area, distribution of targets and offenders in the area, and socio-economic condition of the area. This study proposes tailored measures for each burglary hotspot, considering their unique spatial, temporal, socio-economic, and environmental characteristics, to effectively reduce, control, and prevent burglary incidents in the respective areas.
... While many studies in Nigeria and abroad have assessed different aspects of property crime and offences like burglary, house break-ins, arson, vandalism and shoplifting/shop breakins among others (e.g. Han, et al., 2013;Badiora, et al., 2017;Borg, & Svensson, 2022;Mwangi, et al., 2022), researchers rarely examine crime and offences associated with land management and administration, especially, as it relates to purchase, acquisition, ownership transfer, sales and re-sales as well as mortgage of landed properties. Whereas, land associated crime and offences abound everywhere, and particularly, in Nigeria. ...
Article
Full-text available
In Nigeria, land has become a sensitive matter that threatens the security of the nation. People have acquired land illicitly by forgery, forced eviction, land grabbing, among others criminalities perpetrated by individuals, groups of individuals, cooperate organizations, and public agencies. This study examines patterns of land associated crimes in Obafemi-Owode Local Government Area of Ogun state, Nigeria. Data was collected from 334 respondents in the study area using simple random sampling and analyzed using frequency counts, percentages and likert scale rating. Findings show that all the land associated crimes were above average in the study area with land fraud and exploitation (CRI = 3.25/5.00) as the most common: majority of respondents (38%) had been victims of this particular crime. The major cause of land associated offences was land grabbing (20%) while the key perpetrators were "omo-onile" (25%), land cartels (18%), and self-centered community, elites and political leaders (18%). Among others, findings indicate loss of livelihoods, and increased poverty and hunger (SRI = 3.63/5.00) as the most significant consequence of these offences. The study concluded that land associated crimes have become a widespread occurrence in the study area with far-reaching socioeconomic , environmental and political significance. Policy and future research implications are discussed.
... There is a well-documented "near-repeat" pattern of crime, particularly for property crime (e.g., burglaries), in which there is a spatial and temporal relationship between a crime event and a repeat crime event thereafter (Bernasco, 2008;Groff & Taniguchi, 2019;Schnell, 2024;Short et al., 2009;Townsley et al., 2003). That is, in a short time frame after an initial property crime (e.g., burglary) the risk of that location or nearby locations experiencing another crime victimization is greatly heightened (Borg & Svensson, 2022;Johnson, 2008). Therefore, since the risk of repeat and near repeat victimization in the nearby area is enhanced following a crime event (Schnell, 2024), immediate police outreach alerting residents about this risk and potential of subsequent crime events may encourage residents to protect themselves and their property. ...
Article
Full-text available
Objectives To investigate whether police deployment of door hangers providing crime prevention tips to recently victimized households and nearby households reduces property crime in treated areas. Methods A randomized field experiment was implemented by randomly assigning the intervention across six neighborhoods in Columbia, SC. The test period lasted 184 days, and the intervention effect was analyzed across burglary, theft, and motor vehicle theft crime types. Results T -tests and negative binomial regression models indicated significant decreases across all property crimes and thefts in treated areas. However, there were decreases in burglaries and motor vehicle theft which were not statistically significant. Conclusions We provide experimental evidence on a light-footprint, super-cocooning police intervention targeting property crime. During a staffing crisis, and since Americans are more likely to experience property crime victimization compared to violent crime, police agencies should consider cost-effective and easy-to-implement strategies such as door hanger intervention to reduce property crime.