Applying AI technology and rough set theory for mining association rules to support crime management

and fire-fighting resources allocation

 

Show-Chin Lee

Department of Information Management, National Changhua University of Education, postgraduate

sarashow@ms38.hinet.net

 

Mu-Jung Huang

Department of Information Management, National Changhua University of Education, professor

mjhuang@cc.ncue.edu.tw

 

Abstract

The missions for the police and fire fighters are to protect for public safety and to fight and prevent from fires, respectively. In this dynamic environment, however, there are many potential dangers and uncertain factors that can’t be predicted. In order to improve the total performance, some rules extracted from criminal and fire-fighting records are needed. The purpose of this paper is to mine association rules from a database to support crime management or fire-fighting resources allocation. The mining procedure consists of two essential modules. One is a clustering module based on a neural network, a Self-Organization Map (SOM), which performs grouping tasks on the tremendous number of database records. The another is a rule extraction module applying rough set theory that can extract association rules for each homogeneous cluster and the relationships between different clusters. An example is for illustration.

 

Keywords: Data mining, Self-Organizing MapSOM, Rough set theory, Rule-based knowledge, Crime management