| Abstract [eng] |
In all countries, cities and their suburbs are the most densely populated areas. They are also the places visited by the largest number of tourists and one-day visitors, who inevitably run the risk of becoming victims of crime. It is, therefore, important, not only at national but also at the international level, to know the structure of urban crime and identify urban areas that differ in terms of their criminogenic situation. This requires a geographical approach and regionalisation based on the quantitative data that can offer it. This paper presents the results of a study using big data regarding violent crime, property crime and infringements against public order registered by the police in 2020 in the territories of three major Lithuanian cities and their suburbs (n = 149,239). Events in open spaces were separately addressed. A series of experiments were carried out using several spatial clustering methods. The automatic zoning procedure method that gave the best statistical results was then tested with different combinations of parameters. In each city, seven types of areas of urban crime were identified. Maps of crime areas (regions) were created for each city. The results of the regionalisation have been interpreted from a socio-geographical point of view and conform with previous sociological urban studies. Seven types of areas of crime have been identified, which are present in all the cities studied and, according to a preliminary assessment, roughly correspond to the socio-demographic and urban zones of each city. The maps of crime areas can be applied for crime prevention planning and communication, real estate valuation, strategic urban development planning and other purposes. |
