Application of data mining in identifying and discovering hidden patterns of theft

Crime prevention has always been one of the basic and important issues in human life that have been applied in various ways throughout history. Due to the development of information and communication technologies and the launch of comprehensive information systems in the police force and the registration of criminals’ information in databases, the use of data mining techniques and knowledge discovery to analyze and track down crimes, including theft, is one of the necessities of the Iranian police and the judiciary. The main purpose of this study is to develop and present a data mining model that, by using existing crime databases and data mining tools and algorithms, can identify and detect crime patterns so that the police can predict the occurrence of crime and prevent the occurrence of crimes by more precise control of the forces and their military arrangement in the crime area. In the proposed model, two coupling rules with the former probability algorithm and clustering with the Chi-mean algorithm are used to extract the patterns from a database with more than one hundred thousand theft records. This paper’s novelty can be mentioned as the methodology and the databank volume, consisting of thousands of cases.

Keywords:

data mining , relational rules , crime patterns , data mining process