Context-Aware IoT Models for Detecting Dangerous Scenarios-A Systematic Review

The integration of Internet of Things (IoT) devices and machine learning (ML) algorithms has significantly advanced the monitoring and protection of vulnerable populations, such as the elderly, individuals with disabilities, and those at risk of unexpected threats like abductions or assaults. This review explores a context-aware threat scenario identification framework that utilizes data from smartwatches and smartphones to assess a user’s geolocation, stress levels, and body posture in real time. Stress and location are interpreted using threshold-based methods, while body posture is evaluated through machine learning classification. The study discusses various IoT-based monitoring systems, data integration approaches, and ML techniques used for accurate anomaly detection and risk prediction. It also examines the privacy, energy efficiency, and environmental adaptability challenges associated with deploying such systems. Future directions highlight the role of edge computing, AI ethics, and healthcare integration in improving the responsiveness and ethical deployment of these technologies. The findings underscore the potential of context-aware IoT-ML systems in enhancing safety, provided critical challenges are addressed effectively.