Spatio-Temporal Crime Prediction Model centered on Analysis of Crime Clusters.
Abstract
Crime is defined as “an intentional act or omission in violation of criminal law, and sanctioned
by the state as felony or misdemeanor”. The Misdemeanors are minor crimes that government
punishes by confinement in local jail for a year or less. Police intend to forecast to forecast
number crime, time, place and types of crime to get precaution. In this project spatio-temporal
crime prediction model is produced using time series forecasting. (ARIMA TECHNIQUE).
The model is generated by exploring Jinja road police division crime 2018 data. The
methodology begins with getting clutters with different clustering algorithm and clustering
techniques are compared in land use and the selected clustering algorithm. Then the prediction
is done by use of ARIMA model.
The prediction in time extent, a time series model (ARIMA) is fitted for each month and the
prediction is done for the next twelve (12) months. Therefore, the proposed model will can give
prediction according to time element to assist police officials in planning and tactical
operations.