Examining Deep Learning Architectures for Crime Classification and Prediction

In this paper, a detailed study on crime classification and prediction using deep learning architectures is presented. We examine the effectiveness of deep learning algorithms in this domain and provide recommendations for designing and training deep learning systems for predicting crime areas, using open data from police reports. Having time-series of crime types per location as training data, a comparative study of 10 state-of-the-art methods against 3 different deep learning configurations is conducted. In our experiments with 5 publicly available datasets, we demonstrate that the deep learning-based methods consistently outperform the existing best-performing methods. Moreover, we evaluate the effectiveness of different parameters in the deep learning architectures and give insights for configuring them to achieve improved performance in crime classification and finally crime prediction.

Relevant issues of crime prevention in the sphere of state procurement

Based on the normative acts in force and domestic scientific literature, this article analyzes social relations associated with the implementation of state procurement procedures. The goal is set to examine criminal activity in the sphere of state procurement for outlining the relevant vectors of crime prevention therein. The article identifies the gaps in the current legislation that negatively affect the dynamics of crime in this sphere. Special attention is given to the statistical method that reveals the key characteristics of criminologically significant information, such as circumstances of crime, classification of crime, and cost of crime. The author offers a range of ideas, the realization of which would help to reduce the amount of crime committed in the process of implementation of state order. The structure of this work includes the analysis of the determinants of crime discovered within the framework of studying the normative-legal, investigative-judicial, and domestic doctrinal sources. The novelty of this research lies in outlining the following vectors leaning on the latest information: improvement of current legislation for increasing the effectiveness of the mechanism of determination of formal-legitimate organizations and failure of cash-out transactions, reference to advanced foreign experience, revision of the mechanism of for assessing the activity of authorized budget holders, give due attention to the question of interaction of various regulatory agencies in terms of implementation of their activity.

Unsupervised identification of crime problems from police free-text data

We present a novel exploratory application of unsupervised machine-learning methods to identify clusters of specific crime problems from unstructured modus operandi free-text data within a single administrative crime classification. To illustrate our proposed approach, we analyse police recorded free-text narrative descriptions of residential burglaries occurring over a two-year period in a major metropolitan area of the UK. Results of our analyses demonstrate that topic modelling algorithms are capable of clustering substantively different burglary problems without prior knowledge of such groupings. Subsequently, we describe a prototype dashboard that allows replication of our analytical workflow and could be applied to support operational decision making in the identification of specific crime problems. This approach to grouping distinct types of offences within existing offence categories, we argue, has the potential to support crime analysts in proactively analysing large volumes of modus operandi free-text data—with the ultimate aims of developing a greater understanding of crime problems and supporting the design of tailored crime reduction interventions.

Unsupervised Identification of Crime Problems from Police Free-text Data

We present a novel exploratory application of unsupervised machine-learning methods to identify clusters of specific crime problems from unstructured modus operandi free-text data within a single administrative crime classification. To illustrate our proposed approach, we analyse police recorded free-text narrative descriptions of residential burglaries occurring over a two-year period in a major metropolitan area of the UK. Results of our analyses demonstrate that topic modelling algorithms are capable of clustering substantively different burglary problems without prior knowledge of such groupings. Subsequently, we describe a prototype dashboard that allows replication of our analytical workflow and could be applied to support operational decision making in the identification of specific crime problems. This approach to grouping distinct types of offences within existing offence categories, we argue, has the potential to support crime analysts in proactively analysing large volumes of modus operandi free-text data – with the ultimate aims of developing a greater understanding of crime problems and supporting the design of tailored crime reduction interventions.

Spatially varying relationships between immigration measures and property crime Types in Vancouver Census Tracts, 2016

We empirically test for spatial heterogeneity or local effects of multiple immigration measures on various property crime classification across Vancouver census tracts, 2016. Using spatially referenced property crime data and census data, we use geographically weighted regression to investigate the neighbourhood-level effects of immigration on crime. We find that estimated parameters vary across space, but these local immigration effects do not always vary significantly at the local level. Overall, significant spatial variation in the effects of immigration on property crime is present. These are important for policy and theory. The identification of varied spatial patterns of immigration effects on crime may help explain some of the inconsistent/disparate results found in neighbourhood-level studies on immigration and crime.

Doctrinal basis of crime classification

The article deals with the issues of formation, development and legislative registration of the Institute for the classification of crimes. Scientific and legislative classification of crimes, rules of formal logic as a basis for differentiation of criminal offenses and individualization of criminal responsibility and punishment are investigated. The term “crime classification” has been interpreted, the principles and functions of classification have been defined. The definition of the classification criterion, the content of public danger in the doctrine of criminal law, for which there is a large number of scientific views, is given. The criteria of public danger as a material sign of a crime are defined. According to Article 12 of the Criminal Code of Ukraine, depending on the severity of the crime, the crimes are divided into crimes of low gravity, moderate, serious and especially serious. The legislative classification of crimes was made taking into account the type of punishment (fine and imprisonment), as well as the amount of punishment. Crime classification is inextricably linked to the principles of formal logic and the laws of dialectics. The analysis of the theoretical provisions of the classification of crimes shows that the current legislative definition of crimes of small gravity, medium gravity, grave and especially grave, - is constructed with the non-observance of some basic rules of formal logic concerning the necessity of using in one classification the same grounds and grounds . It should also be noted that the neglect of the typical sanction by the legislator as the sole classification criterion has led to the emergence of an additional formal criterion in the form of a fine, which in itself contradicts the generally accepted doctrinal provisions for constructing the classification.