INVESTIGATING TEMPORAL AND SPATIAL EFFECTS OF URBAN PLANNING VARIABLES ON CRIME RATE: A GWR AND OLS BASED APPROACH
Keywords: Crime mapping, Spatial-temporal Analysis, GWR, OLS, DBSCAN, Land Use Diversity
Abstract. Spatial, temporal, environmental and urban planning variables are some factors effected crime occurrence as a social undesirable phenomenon. In this paper, the effect of urban planning variables (including land use diversity and police station area) and temporal parameters (including daily and weekly time windows) on crime incidence were investigated by doing some spatial-temporal analysis. To tackle this, at first, in order to determine crime clusters, DBSCAN algorithm is utilized. Then by using the Ordinary Least Square (OLS) and Geographically Weighted Regression (GWR) the global effect of urban planning variables on crime spatial clusters are modelled and compared. Finally, the OLS method is used to investigate the effects of temporal variables on crime spatial clusters. The results demonstrate that there is a negative correlation between crime clusters and the police station coverage, and a positive relationship between crime clusters and land-use diversity. Furthermore, increasing the number of police stations in districts with high land-use diversity can result in reducing the crime rate. Also, the result of our temporal analysis shows that the high crime rate was happened at noon and on weekdays. Therefore, increasing the number of police officers at noon and on weekdays in the districts with high land-use diversity is recommended.