CRIME AGAINST WOMEN

Violence against women has persisted throughout history, manifesting in various forms, including psychological distress, physical harm, and sexual assault. This study presents a crime index based on multiple categories that either directly or indirectly contribute to the emergence of criminal behavior. A composite weighted index was developed, consisting of four sub-indexes that focus on Health, Socioeconomic Status, Education, and the Judiciary. The index’s stability and consistency were evaluated through reliability testing. A hybrid model was formulated by integrating multiple linear regression and robust regression techniques, with random forest and stochastic gradient descent serving as meta-regressors. The models were evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The results indicate that the Index exhibits strong reliability, supported by a notably high correlation.