Natural disasters such as floods, earthquakes, and landslides pose significant threats to human life and infrastructure, particularly in regions where traditional monitoring systems fail to provide timely alerts. Existing disaster management approaches rely heavily on manual observation and threshold-based techniques, which often result in delayed responses and reduced accuracy. To overcome these limitations, this paper proposes Disaster Aid Connect, an IoT-based disaster detection and management system integrated with machine learning for real-time prediction and alerting. The system employs sensors such as MPU6050 for vibration detection, water-level sensors for flood monitoring, and DHT11 for temperature and humidity measurement. A NodeMCU microcontroller collects and preprocesses sensor data, which is transmitted to a cloud platform for analysis. A K-Nearest Neighbours (KNN) algorithm is used to classify environmental conditions into disaster-prone or normal states. The system provides real-time alerts, safe zone recommendations, and supports efficient resource allocation. Experimental results demonstrate reliable performance in terms of accuracy, precision, and recall, making the proposed system a scalable and effective solution for disaster preparedness and risk reduction.
D et al. (Thu,) studied this question.