This analysis explores the formation of stable hot spots and the overall shifts of repeat and near repeat residential burglary over time. Data were obtained from a small college town police department. There were 1,513 repeat residential burglaries between January 2003 and December 2009 that occurred at a total of 356 addresses. Based upon past research it is thought that repeat residential burglaries will cluster in time and space creating stable hot spots and that the centrographic measures of the burglaries will remain relatively constant from year to year. The results found support for the formation of stable hot spots but found that the area in which the repeat residential burglaries were occurring increased over time.

Context. A minority of criminals performs a majority of the crimes today. It is known that every criminal or group of offenders to some extent have a particular pattern (modus operandi) how crime is performed. Therefore, computers' computational power can be employed to discover crimes that have the same model and possibly are carried out by the same criminal. The goal of this thesis was to apply the existing Series Finder algorithm to a feature-rich dataset containing data about Swedish residential burglaries. Objectives. The following objectives were achieved to complete this thesis: Modifications performed on an existing Series Finder implementation to fit the Swedish police forces dataset and MatLab code converted to Python. Furthermore, experiment setup designed with appropriate metrics and statistical tests. Finally, modified Series Finder implementation's evaluation performed against both Spatial-Temporal and Random models. Methods. The experimental methodology was chosen in order to achieve the objectives. An initial experiment was performed to find right parameters to use for main experiments. Afterward, a proper investigation with dependent and independent variables was conducted. Results. After the metrics calculations and the statistical tests applications, the accurate picture revealed how each model performed. Series Finder showed better performance than a Random model. However, it had lower performance than the Spatial-Temporal model. The possible causes of one model performing better than another are discussed in analysis and discussion section. Conclusions. After completing objectives and answering research questions, it could be clearly seen how the Series Finder implementation performed against other models. Despite its low performance, Series Finder still showed potential, as presented in future work.

Context. Multiple studies demonstrate that most of the residential burglaries are committed by a few offenders. Statistics collected by the Swedish National Council for Crime Prevention show that the number of residential burglary varies from year to year. But this value normally increases. Besides, around half of all reported burglaries occur in big cities and only some burglaries occur in sparsely-populated areas. Thus, law enforcement agencies need to study possible linked residential burglaries for their investigations. Linking crime-reports is a difficult task and currently there is not a systematic way to do it. Objectives. This study presents an analysis of the different features of the collected residential burglaries by the law enforcement in Sweden. The objective is to study the possibility of linking crimes depending on these features. The characteristics used are residential features, modus operandi, victim features, goods stolen, difference of days and distance between crimes. Methods. To reach the objectives, quasi experiment and repeated measures are used. To obtain the distance between crimes, routes using Google maps are used. Different cluster methods are investigated in order to obtain the best cluster solution for linking residential burglaries. In addition, the study compares different algorithms in order to identify which algorithm offers the best performance in linking crimes. Results. Clustering quality is measured using different methods, Rule of Thumb, the Elbow method and Silhouette. To evaluate these measurements, ANOVA, Tukey and Fisher’s test are used. Silhouette presents the greatest quality level compared to other methods. Other clustering algorithms present similar average Silhouette width, and therefore, similar quality clustering. Results also show that distance, days and residential features are the most important features to link crimes. Conclusions. The clustering suggestion denotes that it is possible to reduce the amount of burglaries cases. This reduction is done by finding linked residential burglaries. Having done the clustering, the results have to be investigated by law enforcement.

Context. According to Swedish National Council for Crime Prevention, there is an increase of 19% in residential burglary crimes in Sweden over the last decade and only 5% of the total crimes reported were actually solved by the law enforcement agencies. In order to solve these cases quickly and efficiently, the law enforcement agencies has to look into the possible linked serial crimes. Many studies have suggested to link crimes based on Modus Operendi and other characteristic. Sometimes crimes which are not possible to travel spatially with in the reported times but have similar Modus Operendi are also grouped as linked crimes. Investigating such crimes could possibly waste the resources of the law enforcement agencies. Objectives. In this study, we investigate the possibility of the usage of travel distance and travel duration between different crime locations while linking the residential burglary crimes. A filtering method has been designed and implemented for filtering the unlinked crimes from the estimated linked crimes by utilizing the distance and duration values. Methods. The objectives in this study are satisfied by conducting an experiment. The travel distance and travel duration values are obtained from various online direction services. The filtering method was first validated on ground truth represented by known linked crime series and then it was used to filter out crimes from the estimated linked crimes. Results. The filtering method had removed a total of 4% unlinked crimes from the estimated linked crime series when the travel mode is considered as driving. Whereas it had removed a total of 23% unlinked crimes from the estimated linked crime series when the travel mode is considered as walking. Also it was found that a burglar can take an average of 900 seconds (15 minutes) for committing a burglary. Conclusions. From this study it is evident that the usage of spatial and temporal values in linking residential burglaries gives effective crime links in a series. Also, the usage of Google Maps for getting distance and duration values can increase the overall performance of the filtering method in linking crimes.

Context. In year 2013 the number of reported residential burglariesin Sweden was 21000, where only 4-5 percent of those actuallygot solved [1]. The Swedish police is trying to improve their way ofworking to increase the number of solved burglaries, this by structuringthe data collection and analysing with computer science methods.Temporal analysis is the key to gure out when crime actually takesplace. Objectives. This thesis study ve dierent methods for analysingthe temporal data of residential burglaries. The temporal analysis isperformed on three time spans: time of day, day of the week and dayof the month. The objective is to evaluate the ve methods in thethree time spans and decide which method is the most suitable foreach of them. Methods. This study includes three experiments testing all ve methodson the three time spans. The experiments focus on comparing theobserved data against the data of burglaries with a known specictime of the crime. In order to test the performance of each method aChi-squared goodness-of-t test was used, as well as a visual comparisonof the produced plots. Results. The results showed that the Aoristic-method was the mostsuitable method to use when analysing temporal data of residentialburglars, if looking at the time of day, day of the week and day ofthe month. Using the methods we also generated plots of the threetemporal distributions, with an R script. Conclusions. We concluded that using the Aoristic-method is themost suitable method to use to generate plots from the temporal data.We also concluded that using this script with the Aoristic-method togenerate plots, would make it possible for the police to resource allocationaccording to when burglaries actually take place.

Under 2016 hade Höganäs kommun den tredje högsta antalet anmälda bostadsinbrott i Sverige med cirka 4 inbrott per 1 000 hushåll, samtidigt som det rapporterades i genomsnitt cirka 1 bostadsinbrott per 1 000 hushåll i resten av landet under samma tidsperiod. För att förändra den rådande lägesbilden för utsattheten av bostadsinbrott i Höganäs startades under 2018 ett samverkansprojekt med hjälp av identifieringsmetoden MärkDNA för att förhindra inbrott och öka tryggheten bland invånarna. Vår studie syftar till att utvärdera samverkansprojektet i Höganäs och deltagande aktörers erfarenhet av projektet. Utvärderingen undersöker även ett urval av invånarna i Höganäs och deras upplevelse och åsikter om samverkansprojektet. Samverkansprojektet fortlöpte under ett års tid med målsättning att minska antalet bostadsinbrott som ökat i kommunen under 2015 fram till mitten av 2016. Syftet med denna utvärderingen är att utvärdera samverkansprojektet och samarbetet mellan kommunen, polisen och leverantörerna av MärkDNA. Utvärderingen presenterar resultatet av projektet och även resultat från studier som gjorts för utvärderingens syfte. Målsättningen var att 4 500 av kommunens 13 500 hushåll skulle delta i samverkansprojektet genom att införskaffa MärkDNA. Resultatet av utvärderingen visar att 9,7 % av målsättningens önskade antal uppnåddes. Deltagande aktörer som medverkat i samverkansprojektet upplevde projektet som verksamt, men att kommunikationen till Höganäs invånare inte gett den spridningen kommunen önskat. I utvärderingens enkätstudie nämner invånarna att projektet varit av stort intresse, men även att bättre kommunikation efterfrågas. Utvärderingen ämnar ge en viktig inblick i hur MärkDNA påverkar kommuner som liknar Höganäs och vilka segment som krävs för ett framgångsrikt projekt med märkning av ägodelar. / In 2016, the municipality of Höganäs had the third highest rate of reported residential burglaries per residents in Sweden with approximately 4 burglaries per 1.000 households. During the same time period an average of 1 residential burglary per 1.000 households was reported in the rest of the country. In order to change the exposure to burglaries in Höganäs a project started in 2018 in the municipality with property marking to prevent burglaries and increase safety among the residents. Our study aims to evaluate the project and the participating operator’s experiences of the project as well as a sample of the population and their opinions about the project in Höganäs. The project extended over a period of one year with the aim of reducing the number of burglaries that increased in the municipality during 2015 to mid-2016. The purpose of this study is to evaluate the project and the collaboration between the municipality, police and suppliers of property marking. The evaluation presents the result of the project and also results from studies made for the purpose of the evaluation. The goal was that 4 500 of the municipality’s 13 500 households would participate in the collaboration project by acquiring property marking. The results from the evaluation shows that 9,7 % of the project’s desired goal of participation was met. Involved operators that was participating in the project experienced the project as effective, but that the communication to Höganäs’s residents did not give the proliferation the municipality desired. The evaluation aims to provide an important insight into how property marking affects municipalities similar to Höganäs and which factors that are required for a successful project with marking of property.

Context: According to Swedish National Council for Crime Prevention, there is an increase in residential burglary crimes by 2% in 2014 compared to 2013and by 19% in the past decade. Law enforcement agencies could only solve three to five percent of crimes reported in 2012. Multiple studies done in the field of crime analysis report that most of the residential burglaries are committed by relatively small number of offenders. Thus, the law enforcement agencies need toinvestigate the possibility of linking crimes into crime series. Objectives: This study presents the computation of a median crime which is the centre most crime in a crime series calculated using the statistical concept of median. This approach is used to calculate the uniqueness of a crime series consisting of linked residential burglaries. The burglaries are characterised using temporal, spatial features and modus operandi. Methods: Quasi experiment with repeated measures is chosen as research method.The burglaries are linked based on their characteristics(features) by building a statistical model using logistic regression algorithm to formulate estimated crime series. The study uses median crime as an approach for computing the uniqueness of linked burglaries. The measure of uniqueness is compared between estimated series and legally verified known series. In addition, the study compares the uniqueness of estimated and known series to randomly selected crimes. The measure of uniqueness is used to know the feasibility of using the formulated estimated series for investigation by the law bodies. Results: Statistical model built for linking crimes achieved an AUC = 0.964,R 2 = 0.770 and Dxy = 0.900 during internal evaluation and achieved AU C =0.916 for predictions on test data set and AUC = 0.85 for predictions on known series data set. The uniqueness measure of estimated series ranges from 0.526to 0.715, and from 0.359 to 0.442 for known series corresponding to differentseries. The uniqueness of randomly selected crimes ranges from 0.522 to 0.726 for estimated series and from 0.636 to 0.743 for known series. The values obtained are analysed and evaluated using Independent two sample t-test, Cohen’s d and kolmogorov-smirnov test. From this analysis, it is evident that the uniqueness measure for estimated series is high compared to the known series and closely matches with randomly selected crimes. The uniqueness of known series is clearly low compared to both the estimated series and randomly selected crimes. Conclusion: The present study concludes that estimated series formulated using the statistical model has high uniqueness measures and needs to be furtherfiltered to be used by the law bodies.