Аңдатпа
The article examines prospects for integrating Internet of Things (IoT) technologies and machine learning (ML) algorithms to create an intelligent air-quality monitoring system. It additionally describes a patient physiological-monitoring module – covering heart rate (HR), HRV metrics (SDNN, RMSSD, LF/HF), respiratory rate, SpO₂, and blood pressure (BP) – integrated into a unified IoT architecture and data-analysis pipeline. A prototype is presented with streaming aggregation of medical and environmental signals, early-warning rules, and a scenario for jointly interpreting air parameters and patient status. The focus is on IoT sensors, real-time data processing methods, and air-pollution forecasting using hybrid ML models, including Random Forest, Gradient Boosting Machines, and Long Short-Term Memory (LSTM) networks. The importance of improving sensor sensitivity and reliability through nanomaterials such as polyaniline, graphene, and carbon nanotubes is emphasized. The article highlights data protection, energy efficiency, and the resilience of scalable IoT systems, as well as their role in reducing environmental risks and supporting the «smart city» concept. It considers pathways for integrating such systems into urban infrastructure, including solutions for automated data analysis and visualization. The article also discusses prospects for deploying intelligent monitoring systems in industrial and residential infrastructure to enhance environmental oversight. Particular attention is paid to developing forecasting models that account for seasonal and climatic variations affecting pollution levels. An interdisciplinary approach that combines advances in IoT, nanotechnology, and ML is underscored as essential for addressing sustainable urban development challenges. The presented results demonstrate high effectiveness and practical applicability for controlling air pollution, improving public health, protecting the environment, and promoting sustainable urban development.