Background: Actuarial risk estimates are considered the gold-standard way to assess whether forensic psychiatry patients are likely to commit prospective criminal offences. However, these risk estimates cannot individually predict the type of criminal offence a patient will subsequently commit, and often simply assess the general likelihood of crime occurring in a group sample. In order to advance the predictive utility of risk assessments, better statistical strategies are required. Aim: To develop a machine learning model to predict the type of criminal offense committed in forensic psychiatry patients, at an individual level. Method: Machine learning algorithms (Random Forest, Elastic Net, SVM), were applied to a representative and diverse sample of 1240 patients in the forensic mental health system. Clinical, historical, and sociodemographic variables were considered as potential predictors and assessed in a data-driven way. Separate models were created for each type of criminal offence, and feature selection methods were used to improve the interpretability and generalizability of our findings. Results: Sexual and violent crimes can be predicted at an individual level with 83.26% sensitivity and 77.42% specificity using only 20 clinical variables. Likewise, nonviolent, and sexual crimes can be individually predicted with 74.60% sensitivity and 80.65% specificity using 30 clinical variables. Conclusion: The current results suggest that machine learning models have accuracy comparable to existing risk assessment tools (AUCs .70-.80). However, unlike existing risk tools, this approach allows for the prediction of cases at an individual level, which is more clinically useful. The accuracy of prospective models is expected to only improve with further refinement. / Thesis / Master of Science (MSc) / Individuals end up in the forensic mental health system when they commit crimes and are found to be not criminality responsible because of a mental disorder. They are released back into the community when deemed to be low risk. However, it is important to consider the accuracy of the method we use to determine risk at the level of an individual person. Currently, we use group average to assess individual risk, which does not work very well. The range of our predictions become so large, that they are virtually meaningless. In other words, the average of a group is meaningless with respect to you. Instead, statistical models can be developed that can make predictions accurately, and at an individual level. Therefore, the current work sought to predict the types of criminal offences committed, among 1240 forensic patients. Making accurate predictions of the crimes people may commit in the future is urgently needed to identify better strategies to prevent these crimes from occurring in the first place. Here, we show that it is possible to predict the type of criminal offense an individual will later commit, using data that is readily available by clinicians. These models perform similarly to the best risk assessment tools available, but unlike these risk assessment tools, can make predictions at an individual level. It is suggested that similar approaches to the ones outlined in this paper could be used to improve risk prediction models, and aid crime prevention strategies.

No description available.

No description available.

Crime is a behavior disorder that is an integrated result of social, economical and environmental factors. In the world today crime analysis is gaining significance and one of the most popular subject is crime prediction. Stakeholders of crime intend to forecast the place, time, number of crimes and crime types to get precautions. With respect to these intentions, in this thesis a spatio-temporal crime prediction model is generated by using time series forecasting with simple spatial disaggregation approach in Geographical Information Systems (GIS). The model is generated by utilizing crime data for the year 2003 in Bah&ccedil / elievler and Merkez &Ccedil / ankaya police precincts. Methodology starts with obtaining clusters with different clustering algorithms. Then clustering methods are compared in terms of land-use and representation to select the most appropriate clustering algorithms. Later crime data is divided into daily apoch, to observe spatio-temporal distribution of crime. In order to predict crime in time dimension a time series model (ARIMA) is fitted for each week day, Then the forecasted crime occurrences in time are disagregated according to spatial crime cluster patterns. Hence the model proposed in this thesis can give crime prediction in both space and time to help police departments in tactical and planning operations.

The purpose of this literature review was to compile recent conducted reviews about the use of artificial intelligence (AI) in crime prediction and prevention. How capable are current AI- technologies in predicting crime? How capable are current AI- technologies in preventing crime? This literature review included 9 reviews. The results suggested that the currently AI technologies in use were capable of predicting and preventing crime. They could find patterns in large data sets in much more efficient manner than humans could. The most commonly used AI technologies in crime prediction were data mining, and machine learning and deep learning were most commonly used in crime prevention. The results of this literature review indicated that the use of AI systems in crime prediction and prevention is growing, however it can still be viewed as a new phenomenon that requires further research. To achieve higher levels of accuracy there was requirements for access to larger data sets and further testing and training of the AI models. Criminologists also needs to pay more attention to the field of AI in crime prediction and prevention so that they can direct the practice.

The low amount solved residential burglary crimes calls for new and innovative methods in the prevention and investigation of the cases. There were 22 600 reported residential burglaries in Sweden 2017 but only four to five percent of these will ever be solved. There are many initiatives in both Sweden and abroad for decreasing the amount of occurring residential burglaries and one of the areas that are being tested is the use of prediction methods for more efficient preventive actions. This thesis is an investigation of a potential method of prediction by using neural networks to identify areas that have a higher risk of burglaries on a daily basis. The model use reported burglaries to learn patterns in both space and time. The rationale for the existence of patterns is based on near repeat theories in criminology which states that after a burglary both the burgled victim and an area around that victim has an increased risk of additional burglaries. The work has been conducted in cooperation with the Swedish Police authority. The machine learning is implemented with convolutional long short-term memory (LSTM) neural networks with max pooling in three dimensions that learn from ten years of residential burglary data (2007-2016) in a study area in Stockholm, Sweden. The model's accuracy is measured by performing predictions of burglaries during 2017 on a daily basis. It classifies cells in a 36x36 grid with 600 meter square grid cells as areas with elevated risk or not. By classifying 4% of all grid cells during the year as risk areas, 43% of all burglaries are correctly predicted. The performance of the model could potentially be improved by further configuration of the parameters of the neural network, along with a use of more data with factors that are correlated to burglaries, for instance weather. Consequently, further work in these areas could increase the accuracy. The conclusion is that neural networks or machine learning in general could be a powerful and innovative tool for the Swedish Police authority to predict and moreover prevent certain crime. This thesis serves as a first prototype of how such a system could be implemented and used.

Diverse teorie si concentrano sui legami tra criminalità e caratteristiche specifiche di luoghi e comunità. Tuttavia, solo pochi studi applicati sostengono esplicitamente che i fattori contestuali possono combinarsi nel determinare il rischio di criminalità e che le loro influenze criminogene possono operare su scala diversa. Questo studio si propone di indagare come alcune caratteristiche del paesaggio urbano (microlivello) interagiscono tra loro, nonché con le caratteristiche demografiche, economiche e sociali dell'ambiente dei quartieri circostanti (livello meso), per determinare la vulnerabilità spaziale alla criminalità e, in definitiva, la probabilità di un evento criminale. Questo studio conduce una valutazione del rischio di criminalità spaziale per rapine e crimini violenti in due grandi aree urbane: Milano, Italia e Iztapalapa, Messico. I casi di studio sono focalizzati su due paesi molto diversi, il che consente sia la valutazione dell'influenza di effetti contestuali più ampi (livello macro) sia la verifica di alcuni presupposti teorici al di fuori dell'ambiente anglosassone. L'analisi si fonda sull'approccio del Risk Terrain Modeling. Tuttavia, contrariamente alle applicazioni precedenti, l'analisi in questo studio si basa su un modello di regressione multilivello che include termini di interazione. Lo studio propone inoltre metodi innovativi attraverso i quali esporre e comunicare i propri risultati. Nel complesso, i risultati dimostrano che fattori contestuali misurati a diverse scale geografiche interagiscono in modo significativo tra loro per determinare il rischio di criminalità. Questa scoperta suggerisce di combinare input provenienti da diverse teorie al fine di comprendere le dinamiche alla base del verificarsi del crimine. Inoltre, il metodo proposto generalmente consente di prevedere meglio i crimini futuri e consente la generazione di narrazioni di rischio più precise per informare politiche e interventi. / Several theories focus on the links between crime and specific characteristics of places and communities. However, only a few applied studies explicitly purport that contextual factors may combine in determining crime risk and that their criminogenic influences may operate at different geographical scales. This study aims to investigate how certain features of the urban landscape (micro-level) interact with each other, as well as with demographic, economic and social characteristics of the surrounding neighbourhoods (meso-level), to determine spatial vulnerability to crime and, ultimately, the likelihood of a criminal event. This study conducts a spatial crime risk assessment for robberies and violent crimes in two large urban areas: Milan, Italy and Iztapalapa, Mexico. The case studies are focused on two very different countries, which allows for both the assessment of the influence of broader contextual effects (macro-level) and to test certain theoretical assumptions outside the Anglo-Saxon environment. The analysis is grounded in the Risk Terrain Modeling approach. However, in contrast to previous applications, the analysis in this study relies on a multi-level regression model including interaction terms. The study also proposes innovative methods through which to display and communicate its findings. Overall, the results demonstrate that contextual factors measured at different geographical scales interact significantly among them to determine crime risk. This finding suggests combining inputs from different theories in order to understand the dynamics behind crime occurrence. Furthermore, the proposed method generally allows us to better predict the locations of future crimes and enables the generation of more precise risk narratives to inform policies and interventions.