ABSTRACT
This study aims to construct a predictive analytics system leveraging temporal and locational
data to forecast crime occurrences, identify crime patterns, and enhance strategic law
enforcement. Covering a period from its inception to May 2024, the methodology employed
includes the use of ARIMA and LSTM models for time series forecasting. The system
integrates diverse data sources, ensuring comprehensive crime analysis. Results indicate a
significant improvement in crime prediction accuracy, enabling proactive crime prevention
and contributing to the knowledge of crime data analysis by incorporating real-time data and
advanced machine learning techniques.
