Identification of crime-prone areas and effective factors in its escalation using spatial data mining methods (case study: Isfahan city)

Moosavi, Bahareh Sadat; Toomanian, Ara; Argany, Meysam; Arvandi, Samane; Saedpanah, Mehran

doi:10.22059/jurbangeo.2023.350320.1749

Identification of crime-prone areas and effective factors in its escalation using spatial data mining methods (case study: Isfahan city)

Document Type : Research article

Authors

Bahareh sadat Moosavi ¹

Ara Toomanian ²

Meysam Argany ³

samane arvandi ⁴

Mehran Saedpanah ⁵

¹ PhD Candidate of Geographical Information System and Remote Sensing, Department of Physical Geography, Faculty of Geography, University of Tehran, Tehran, Iran

² Dr. Ara Toomanian Associate Professor Department of GIS &amp;amp;amp; RS, Faculty of Geography, University of Tehran, Tehran, Iran

³ Department of Remote Sensing and GIS, Faculty of Geography, University of Tehran

⁴ معاونت آب و خاک، وزارت جهاد کشاورزی

⁵ university college omran & tosee

10.22059/jurbangeo.2023.350320.1749

Abstract

The findings of the associational rule extraction revealed a spatial relationship between the distance from the police station and the usage of opiates and hashish, which has a direct relationship on the use of the park. Additionally, there is a direct correlation between the use of marijuana and grass and police stations, indicating that these drugs have been used close to these institutions. The majority of hashish, opium, and grass drug use occur in close proximity to roads, parks, and schools. Additionally, compared to other substances, marijuana and grass have a direct spatial relationship with the police station, and use of these drugs can be witnessed close to the station. Despite being extremely prevalent in the database, it appears that the opium drug did not significantly correlate spatially with the chosen uses. Finally, the top 10 laws were determined and their effects on confidence, support, and lifting were examined. In the context of a spatial information system, the usage of spatial distribution functions can be useful for pinpointing crime hotspots and assisting law enforcement with management decisions. With its research of crime trends and crime statistics, it can assist the police and other relevant agencies in locating potential crime hotspots and preventing them in the future. Additionally, it might be highly helpful to apply data mining techniques to anticipate and prevent crimes as well as their locations.

Keywords

Association Rules

Data Mining

Drugs

Hotspots

Kernel Density Estimation Method