Mapping the Risk Terrain for Crime Using Machine Learning

Abstract

Objectives: We illustrate how a machine learning algorithm, Random Forests, can provide accurate long-term predictions of crime at micro places relative to other popular techniques. We also show how recent advances in model summaries can help to open the 'black box' of Random Forests, considerably improving their interpretability. Methods: We generate long-term crime forecasts for robberies in Dallas at 200 by 200 feet grid cells that allow spatially varying associations of crime generators and demographic factors across the study area. We then show how using interpretable model summaries facilitate understanding the model's inner workings. Results: We find that Random Forests greatly outperform Risk Terrain Models and Kernel Density Estimation in terms of forecasting future crimes using different measures of predictive accuracy, but only slightly outperform using prior counts of crime. We find different factors that predict crime are highly non-linear and vary over space. Conclusions: We show how using black-box machine learning models can provide accurate micro placed based crime predictions, but still be interpreted in a manner that fosters understanding of why a place is predicted to be risky.

Redrawing Hot Spots of Crime in Dallas, Texas

“…Subsequently a limitation of this work is that we do not consider here other methods used to identify hot spots and test their accuracy. Many exist, in addition to the traditional clustering techniques mentioned in the literature review, there are a variety of model based approaches (Drawve, 2016; Mohler et al., 2018; Wheeler & Steenbeek, 2020) we have not touched on here. Thus while we cannot say that our DBSCAN analysis is dispositive that it is better than these other approaches, we believe it is likely the case one can use many of these different approaches to improve upon the current Dallas TAAGs, even simply just counting up street segments with the highest crimes is likely much more accurate (MacBeth & Ariel, 2019; Wheeler & Steenbeek, 2020).…”

“…We eliminated several lochs from the city outline of Dallas proper, as including these areas would artificially increase PAI statistics, although they cannot reasonably have geocoded crime incidents occur within them (Wheeler & Steenbeek, 2020). Figure 1 displays a map of Dallas proper, along with the 2017 Dallas PD identified TAAG areas.…”

“…These parameters are chosen here as 400 feet for epsilon, and $400,000 for the total costs of crime within a hotspot. Four hundred feet was chosen as this is slightly less than the average street segment size in Dallas (Wheeler & Steenbeek, 2020). So for an average length street segment in Dallas, if there were only crimes reported at either end, this technique would not link them together.…”

Breaking the Chain: How Arrests Reduce the Probability of Near Repeat Crimes

“…Given these are control variables and are not the main interest of the analysis, we do not spend substantial time examining these factors in the subsequent models. Wheeler and Steenbeek (2020) provide a more detailed description of the data sources and how they are relevant in predicting microlevel crime patterns, which similarly extends to predicting near repeat crime patterns (Garnier et al, 2018; Piza & Carter, 2018). These particular crime generator factors include commercial establishments: (1) large business retailers (e.g., Walmart, The Home Depot, CVS), (2) smaller food and clothing stores, (3) gasoline stations, (4) eating and drinking places, (5) liquor stores, (6) large entertainment areas (e.g., movie theaters, concert halls), (7) smaller entertainment areas (e.g., gyms, bowling alleys) and hair salons, (8) motels, (9) hotels, (10) shopping malls, (11) banks, and (12) check cashing stores.…”

Cyber-attack method and perpetrator prediction using machine learning algorithms

“…Other areas of use are to determine spam and network attacks ( Canbek, Sagiroglu & Temizel, 2018 ), to detect the phishing attacks against the banking sector ( Moorthy & Pabitha, 2020 ) and to reduce sexual crimes on social media ( Ngejane et al, 2018 ). These methods have been implemented in fields as stock prediction ( Gurjar et al, 2018 ), risk mapping by crimes ( Wheeler & Steenbeek, 2020 ) and cyber profiling ( Zulfadhilah, Prayudi & Riadi, 2016 ). Predicting crime trend and pattern ( Biswas & Basak, 2019 ), criminal identity detection ( Bharathi, Indrani & Prabakar, 2017 ) and crime prevention ( Lin, Chen & Yu, 2017 ) are also areas of implementation.…”