FireDS-IoT: A Fire Detection System for Smart Home Based on IoT Data Analytics

Fire is the most widespread cause of death by accident. Fire affects thousand of residents each year, resulting in injury and loss of life. In this paper, an Internet of Things (IoT) based Fire Detection System (FireDS-IoT) is designed to prevent people from fire by providing an alert message in the emergency. The system is designed using MQ-135 (CO_2), MQ-2 (smog), MQ-7 (CO) and DHT-11 (temperature) sensors embedded with Arduino to get the fire event information in the surrounding more accurately. This research distinguishes the conditions in a surrounding as fire, no fire, and may be fire. This classification is performed using the K-Nearest Neighbors (K-NN) and decision tree machine learning algorithms in Python. Several scenarios were recorded in the experiment for training. Results show that K-NN and decision tree shows an accuracy of 93.15% and 89.25%, respectively. As a result, we were able to prove that K-NN provides more accuracy in detecting fire. Therefore, it is used for classification, and if fire conditions arise then a safety message is sent to the registered mobile number using Python programming.