Towards a Crime Hotspot Detection Framework for Patrol Planning

Araujo, Adelson de; Cacho, Nélio; Bezerra, Leonardo C. T.; Vieira, Carlos; Borges, Julio

doi:10.1109/hpcc/smartcity/dss.2018.00211

Cited by 22 publications

(13 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although we have found many different forms of data aggregation into features, both spatially and temporally, the procedure of assigning features is often insufficiently reported, making it difficult to reproduce the proposed methodology. Still, well-defined workflows or frameworks followed by feature-engineering dependent methods were detailed in Malik et al (2014) and Araújo et al (2018). They synthesized their forecasting methods in (1) aggregate raw data spatially, following a crime mapping methodology (e.g., counting events inside grid cells), (2) generate time series and their features, (3) fit a forecasting model using an algorithm, and (4) visualize the results.…”

Section: Proposed Methods Inputmentioning

confidence: 99%

“…The third category of forecasting algorithms, the traditional ML, is split up almost equally between classification and regression tasks. Only three articles discussing traditional ML algorithms do not mention information about the baseline comparison (Araújo et al 2018;Rodríguez et al 2017;Rummens et al 2017). The majority of ML algorithms (n = 11) use the training-testing split validation strategy applied to the classification task.…”

Section: Algorithms and Validation Strategiesmentioning

confidence: 99%

“…Although most papers use standard evaluation metrics, such as PA for a binary outcome, they usually do not include complementary metrics, in order to ensure that every aspect of the prediction performance is covered. Often, the PA is used by itself to measure model performance (Araújo et al 2018;Malik et al 2014;Mu et al 2011). Complementary metrics are needed, because whilst one method may have a higher evaluation score than others, they may provide additional information.…”

Section: Dependencies and Limitationsmentioning

confidence: 99%

See 2 more Smart Citations

A systematic review on spatial crime forecasting

et al. 2020

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Section: Proposed Methods Inputmentioning

confidence: 99%

Section: Algorithms and Validation Strategiesmentioning

confidence: 99%

Section: Dependencies and Limitationsmentioning

confidence: 99%

See 1 more Smart Citation

A systematic review on spatial crime forecasting

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly, a measure termed 'precision'-defined as the proportion of crimes that were correctly predicted by the model out of the total number of crimes predicted by the model (TP/(TP + FP))-has also been proposed [6,44]. Finally, a measure termed 'predictive accuracy' (PA)-which measures the proportion of crimes correctly classified out of the total number of crimes ((TP + TN)/(TP + FP + TN + FN))-has also been used [45][46][47].…”

Section: Measures For Comparing Crime Modelsmentioning

confidence: 99%

Considerations for Developing Predictive Spatial Models of Crime and New Methods for Measuring Their Accuracy

Joshi

Curtis‐Ham²,

D'Ath

et al. 2021

IJGI

View full text Add to dashboard Cite

A literature review of the important trends in predictive crime modeling and the existing measures of accuracy was undertaken. It highlighted the need for a robust, comprehensive and independent evaluation and the need to include complementary measures for a more complete assessment. We develop a new measure called the penalized predictive accuracy index (PPAI), propose the use of the expected utility function to combine multiple measures and the use of the average logarithmic score, which measures accuracy differently than existing measures. The measures are illustrated using hypothetical examples. We illustrate how PPAI could identify the best model for a given problem, as well as how the expected utility measure can be used to combine different measures in a way that is the most appropriate for the problem at hand. It is important to develop measures that empower the practitioner with the ability to input the choices and preferences that are most appropriate for the problem at hand and to combine multiple measures. The measures proposed here go some way towards providing this ability. Further development along these lines is needed.

show abstract

“…Araujo et al [30] highlighted the use of an algorithmically an infrastructure architecture, which can then be implemented to service current and future public protection networks and channels. The system is based on the ROTA strategy, with public safety initiatives as its background, which was devised for the greater purpose of improving urban services in Brazil's new African metropolis, Natal.…”

Section: Crime Hotspot Predictionmentioning

confidence: 99%

Geospatial Crime Analytics: A GIS-Based Approach Towards Prediction of Crime Hotspots

Panigrahy¹

2021

Preprint

View full text Add to dashboard Cite

GIS has proven its proficiency in many fields but still require some more observations. The spatial interpolation which has seen widespread used in diverse domains with multiple applicability. Currently, some commercial GIS or analytics program packages now provide spatial interpolation technique including inverse distance weighting (IDW), kriging, triangulated irregular networks (TIN) interpolation, kernel density interpolation (KDE). Further, the two spa-tial autocorrelation techniques i.e., also known as LISA indicators that comprised of Moran’s I and Getis-Ord Gi* techniques. Amalgamation of spatial interpolation and spatial autocorrelation are being shown in this paper. For theimplementation, related enabling technologies are summarized. Further, this paper entailed the literature survey of multiple papers comprising the associated works carried out in the domain of spatial interpolation and spatial autocorre-lation areas of GIS. The proposed framework is a generalized framework with four tiers to realize geospatial systems. This paper scrutinized geospatial crime dataset for understanding the proposed framework realistically. Numerous results are produced to prove the efficiency of the proposed model. The paper entailed weighted-overlay analysis, symbological heatmap analysis, choropleth analysis, kernel density interpolation, and Getis-Ord Gi* spatial autocorrelation analysis implemented to practically realize the essence of geospatial crime analytics. Finally, future research directions are recapitulated, where the task of interpolation and autocorrelation might be merged with techniques of artificialintelligence, machine learning, and cloud computing paradigm, serverless com-puting framework and many such in order to enhance the proposed structurewith lessened restrictions.

show abstract

Towards a Crime Hotspot Detection Framework for Patrol Planning

Cited by 22 publications

References 26 publications

A systematic review on spatial crime forecasting

A systematic review on spatial crime forecasting

Considerations for Developing Predictive Spatial Models of Crime and New Methods for Measuring Their Accuracy

Geospatial Crime Analytics: A GIS-Based Approach Towards Prediction of Crime Hotspots

Contact Info

Product

Resources

About