Crime Prevention and Addressing Violence Against Women

Abstract

The safety of women remains a pressing societal concern, with many facing threats like harassment, rape, molestation, and domestic abuse due to various sociocultural factors. The Internet of Things (IoT) has emerged as a promising tool to address these issues. This study systematically reviews research papers on IoT-based devices for women's safety, analyzing key features, wearable components, sensor types, and machine learning algorithms used. The review covers articles published between 2016 and 2022. It finds that pulse-rate and pressure sensors are commonly used to monitor women in distress, while technologies like GPS, GSM, and Raspberry Pi enable alert transmission. Machine learning algorithms such as logistic regression, hidden Markov models, and decision trees help identify women at risk and prevent dangerous situations. The review also highlights the need for improved systems that focus on automatic alert generation with minimal human interaction and greater accuracy. In addition, the study proposes a taxonomy categorizing various techniques, features, and sensors, along with an architectural model for developing IoT-based safety devices. Finally, it underscores the importance of integrating multiple sensors to enhance threat detection accuracy, while identifying gaps and challenges in practical applications.