Modified ARIMA Model for Improving Certainty in Spatio-Temporal Crime Event Prediction

Crimes are serious threat to society. The recent economic developments and globalizations have bloated it as serious global concern associated with demographic factors due to which crime site selection and pattern identification have became a central concern in crime prediction. During spatio-temporal crime prediction Linear Model and Generalized Linear Model are used for quantitative improvement in prediction. Both of these models lack certainty in crime prediction. In this paper we have enhanced the Generalized Linear Model for Crime Site Selection and analyze it for crime events using Modified ARIMA (Auto Regressive Integrated Moving Average) with big data technologies. Such enhancement is support similar crime trends among various crime locations for criminal site selection. The simulation results show that out model presents a more significant insight into the scope and complexion of crime prediction and improves certainty in crime prediction.