| Abstract [eng] |
Spatial data provides very important insights into the spatial patterns of criminal activity, allowing identify high-risk areas, understand crime trends, to assess the spatial relationships between crime and various social, economic and environmental factors. Using spatial data, decision makers can prepare targeted intervention measures, allocate resources effectively and implement proactive measures aimed at crime in cities and in the suburbs, which successfully identifies more efficient and sustainable solutions. to increase public safety and community well-being. This study analyses incidents recorded by the Police that have the characteristics of criminal offences and are one of the best indicators of crime. The events analysed are violence-related incidents recorded between 2015 and 2022; drug-related incidents; public order offences and theft and destruction/damage to property. Investigating crime in major cities and their suburbs requires a multifaceted approach, from defining what constitutes a suburb to developing methods for precise identification of suburban areas. Suburbs are generally characterized by their proximity to urban centers, lower population density compared to cities, and residential and commercial land use. We use a variety of spatial data layers, including address points, population data, road networks and building footprints, to pinpoint suburbs. By analyzing these data sets and applying spatial analysis techniques such as density and proportion calculations, suburban boundaries can be determined. It was noted that there are no studies that use spatial analysis methods and geographic data to identify the suburbs of large cities. This study focuses much more on the spatial distribution and regionalization of crime, and the methodology for identifying suburbs needs to be improved. Once suburbs are delineated, conducting spatial analysis of crime within these areas requires diverse methods to capture the complexity of the crime patterns. Three commonly employed methods include density maps, densities differences, and location quotients were applied. We found out that in the major cities, the majority of incidents are recorded in the central parts of the cities; violent incidents are more frequent in residential areas around city centers; the intensity of public order offences is highest in the central parts of cities; property crimes are highest in central cities parts with hot spots near shopping centers and main roads. Based on 2020 data, more than 500 experiments have been carried out using different spatial clustering methods and parameter combinations. The automatic zoning procedure method, which gave the best statistical results, was then tested using different combinations of parameters. Seven types of urban crime zones were identified in each city. Maps of the crime zones (neighbourhoods) in each city were drawn. The results of the zoning were interpreted from a socio-geographical perspective and found to be at least partially consistent with previous sociological studies of cities. Seven types of crime areas 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 areas of each city. The crime area maps can be used for crime prevention planning and communication, real estate valuation, strategic urban development planning and other purposes. |
