Spatiotemporal Frequent Pattern Mining : A Case Study in Crime Pattern Analysis

Abstract

Spatiotemporal pattern mining tried to discover unknown potentially interesting and useful event sequences where events occur within a specific time interval and locate geographic close to each others Previous works use partition or ill-defined representation of spatial objects and neglect some spatial properties exist in original spatiotemporal data Moreover traditional sequential pattern mining methods don't suit the non-transactional spatialtemporal database In this paper we expose the disappearance of spatial correlation due to improper data representation and propose a naive approach to mine frequent sequential spatiotemporal pattern The end of the paper is a case study of crime pattern analysis

Date of Award	2014 Aug 27
Original language	English
Supervisor	Kun-Ta Chuang (Supervisor)