Authors - Usman Ali, Tariq Hussain, Muhammad Mujeeb Akbar, Shahzad Ali, Sajid Abstract - In Pakistan, crime is always a significant economic and social issue affecting governance and public safety. Religious issues, poverty, and political instability have a very high contribution to increasing crime in the country. Despite highly effective efforts of authorities and prevention techniques, crimes are often inconsistent and lack effectiveness in determining crimes across the country. This study aims to analyze national crimes in Pakistan using machine learning and structured datasets to support more effective decision-making and understanding of crime rates across the country. Through the proposed framework, linear regression, the Decision Tree, and the Random Forest are applied to historical data obtained from the Kaggle repository. With different evaluation matrices and cost functions, the proposed model is interpreted and evaluated. The Evaluation metric includes R², and the cost functions include MSE, MAE, and RMSE. Linear Regression works effectively in comparison to others, providing the highest R², i.e., 99.9%. By using this framework, crimes from 2030 to 2050 are predicted with a difference of 5 years. These algorithms help to identify different types of crimes across the country based on different factors. Unlike traditional methods and black-box models, this research underscores the comprehensibility, making insights understandable for both law and policy makers. The study explains how interpretable frameworks can help authorities to find high-risk area predictions. In the future, the aim is to use deep learning techniques with explainable AI (XAI) for enhancement in this study, which can support the development of more public safety systems across the country.
