Mining Regular Frequent Crime Patterns using Vertical Format

Background/Objectives: The goal of crime data mining is to understand various crime patterns in criminal behavior in order to predict crimes and anticipate criminal activity to avoid the crime not to happen. Methods/Statistical Analysis: Predicting crime is one of the global challenges facing by Law enforcement department and it requires persistent efforts in order to restrict. In this paper we are introducing a new crime pattern called regular frequent crime pattern which occurs regularly at certain time intervals using vertical data format also satisfies downward closure property. Findings: Crime patterns were not defined by statistics and its identification is more than just counting and summarizing crimes that are similar in characteristics and/or location on a map. Crime pattern is a group of one or more crimes reported to or discovered by the police. The approaches which are pattern based have the possibility to help the police department in discovering new type of crime patterns. Applications/Improvement: Our experiment results show the impact on execution time and memory. This project is also useful for police department in finding the regular-frequent crimes which are happening in today's world.