Sexual Violence Risk Assessment and Prediction in Indian States using Machine Learning

Sexual violence remains one of the most pressing human rights concerns in contemporary society, particularly in India where rising rape incidents highlight systemic vulnerabilities in safety, justice and prevention mechanisms.Despite the enactment of stringent legal frameworks such as the Nirbhaya Act and numerous awareness campaigns, the persistence of rape cases points to deeper sociocultural, economic, and institutional failings.In response to this critical issue, this study introduces a data-driven solution leveraging machine learning (ML) to proactively assess and mitigate rape risks across Indian states.The model integrates demographic parameters like age, geographic indicators such as state location, and historical crime data to classify locations into risk categories.Three ML algorithms-Logistic Regression, Support Vector Machine (SVM) and Random Forest are employed to evaluate predictive performance.The Random Forest Classifier demonstrated an accuracy of 90% success rate in identifying high-risk zones.This approach aims to support law enforcement agencies, policymakers, and public safety planners in targeted intervention strategies, offering a scalable and intelligent framework to enhance women's safety in India.