Efficient frequent pattern knowledge for crime situation recognition in developing countries

Abstract

Public safety of citizens, as well as crime prevention and control, are issues of great concern to the government and public safety agencies. With the existence of crime databases in the society, data mining techniques are viable tools for crime situation recognition and control. However, most existing techniques often fall short to consider relatively new itemsets during the mining process. Our research focus is to create a flexible and effective, yet simple, solution to crime situation recognition. Building upon the Fp-Growth algorithm, we present CitiSafe as a tool, with data mining in the back-end and visualisation in the front-end. The tool can assist law enforcement agencies and public safety organisations to channelize their resources accordingly to achieve a focused and effective crime prevention strategy i.e decide patrol boundaries and plan patrol routes.