SafeGuardHer: Blockchain and Ensemble Learning based CPS Framework for Women’s Safety Using Wearable Devices

This paper presents a comprehensive framework of machine learning and blockchain to bolster women’s safety through the integration of wearable technology. Human wearables in this framework include bracelets and earrings worn by people and embedded with sensors to detect physiological and environmental parameters: for example, the pulse rate, body temperature, and noise in the environment. These parameters are analyzed using the Light Gradient Boosting Machine (LightGBM), an ensemble learning model is chosen to give excellent performance in predictive accuracy and computational efficiency. From the result, it can be observed that the LightGBM model is better compared to other comparative machine learning models. It has an accuracy of 94%, with precision and recall rates at 91.5% and 95.6%, respectively. These indicate that the model based on these measures is a good predictor of the stress indicator, a pivotal point of signaling timely alerts and interventions in unsafe conditions. Integrating such devices with blockchain technology further adds to the framework’s utility by allowing real-time communication and location tracking while also ensuring the integrity and security of sensitive data. Hence, this tool enhances women’s safety in diversified environmental settings. This synthesis of wearable technology with advanced data analytics also called Cyber Physical System (CPS) is a game-changer in devising proactive safety measures for women.