Advanced IoT and ML-driven personal safety technology for women

Abstract

This article introduces VigilNet, a personal safety system powered by internet of things (IoT) and machine learning (ML) that identifies environmental irregularities instantaneously through intelligent wearable gadgets. In contrast to solutions dependent on extensive datasets and reactive alert systems VigilNet employs a few-shot learning (FSL) strategy utilizing siamese neural networks (SNNs) to achieve precise anomaly recognition, with limited training data. In the information and communication technology (ICT) realm, VigilNet is a smart IoT sensing and communication network with interconnected wearable sensor nodes continuously collecting physiological and environmental data for real-time processing. The SNNs based FSL module then fuses these multiple modal streams of sensor information and extracts the similarity-based threat assessment, which in turn drives an intelligent decision support system to alarm the operator to take timely action for any abnormal situations. VigilNet, thus, combines IoT connectivity, real-time data analytics, and machine-learning-driven decision support into a single personal safety system. The framework was tested on an IoT wearable dataset encompassing heart rate, body temperature, ambient noise and proximity data. Experimental results demonstrate that VigilNet achieves a threat detection rate (TDR) of 0.85, false alarm rate (FAR) of 0.10, and a response time (RT) of 3.5 seconds, outperforming conventional IoT–ML safety models. These findings validate the feasibility, efficiency, and robustness of the proposed architecture and confirm the potential of FSL-based techniques for next-generation women’s safety technologies.