Data classification for air quality on wireless sensor network monitoring system using decision tree algorithm

Currently, the air quality monitoring becomes important things for knowing the value of air pollution especially in the cities. In our previous research, we built the air quality monitoring system using wireless sensor network (WSN). Each sensor nodes will transmit all of the air quality data to the base station controller (BSC). This data consists of weather condition, temperature, humidity, carbon monoxide (CO) and carbon dioxide (CO2). In this paper, we propose an implementation of the classification algorithm for classifying air quality in BSC node for generating the danger status of the warning system. By using the C4.5 algorithm, the entropy and information gain values of each case were determined in order to construct the decision tree structure and building the rule sets. From the experiment using confusion matrix, we can see that the proposed decision tree algorithm has the capability to classify the air quality data from sensor nodes with the accuracy of 85.71%, the precision of 81.82%, the sensitivity of 60.00%, and the specificity of 92.31%.