The GIS based Criminal Hotspot Analysis using DBSCAN Technique

Abstract

Spatially Data mining used efficiently to extract any potential patterns and associations to detect hidden information from multiple sources data. In this paper, data mining Density-based spatial clustering of applications with noise DBSCAN algorithm is emphasised. The importance in this work was using a prototype software to process the giving data into an understandable outcome throw clustering technique, it is a powerful method for criminal activities detection and pattern recognition to get useful information that can help police to reduce crimes. Spatial data mining is practical with geographical crimes data set and processing a large amount of crimes data. Police conventional way was manual and time-consuming using a pin on the wall. Therefore, it has to be developed and merged with advanced techniques. In this study, data mining clustering method was used to examine Baltimore, Maryland's crimes information. The processed criminal data from the state of Maryland, Baltimore City was 340,924 cases and 16 attributes to reflect the cases between 2012-2018. DBSCAN algorithm is utilized to cluster crimes incidents focused on certain predefined events and the outcome of these clusters employed to find hotspots. The clustering findings are visualized by the GIS to make crimes distribution on the map at real-time for the law enforcement to understand and interact