Crime location prediction based on the maximum-likelihood theory

Abstract

In this paper, we are required to construct models to predict the offender's location. Initially, we take a rough anchor point locating based on psychological analysis of offenders. Then we construct a mathematical model based on the maximum-likelihood theory to make accurate predictions about the anchor point and next crime locations. Then we choose an actual crime cases to test the reliability and utility of our model. Finally, a conclusion that only when those crime sites are evenly distributed around a point can be drawn and accurate mathematical predictions can be realized. © 2011 Springer-Verlag.