Artificial Intelligence for Safer Cities: A Deep Dive into Crime Prediction and Gun Violence Detection

Abstract

Crime poses a significant threat to national security, necessitating efficient crime analysis methods based on spatial and temporal data. Traditional techniques like paperwork and statistical analysis are inadequate for accurate crime prediction. However, the integration of machine learning and data mining has significantly enhanced crime analysis accuracy. This study comprehensively surveys various criminal analysis and prediction methods employing machine learning and data mining tech- niques, aiming to provide a concise overview of their application in crime prediction. The review offers valuable  insights  to  crime researchers and supports future research by addressing crime definition challenges, prediction system complexities, and classifications through comparative studies. Literature demon- strates that supervised learning, particularly Logistic Regression, has been widely utilized for crime prediction, showcasing its effectiveness in this domain.