MACHINE LEARNING APPROACH TO PREDICT AND COMPARE THE AIR QUALITY INDEX IN A CONFINED ENVIRONMENT

Indoor air pollution is very dangerous as people spend the majority time indoors. Cooking areas are found to be hazardous as there would be an emission of harmful pollutants. This is due to the continuous cooking process which affects people working there causing them various diseases, especially carbon monoxide poisoning. The purpose of this research is to evaluate several machine learning algorithms like support vector machine (SVM), K-nearest neighbor (KNN), logistic regression (LR), and decision tree (DT) for predicting the air quality index (AQI) of a Barbeque Nation Hotel kitchen's confined interior environment. This investigation was done based on real-time data that was gathered by an indoor air quality monitoring system which was placed inside the kitchen for a few weeks under various cooking conditions. Results show that DT has the highest accuracy of 98.79% followed by KNN with an accuracy of 93.01%. SVM has an accuracy of 80.34%, and LR has a low accuracy of 80.20%. Therefore, DT which is a classification algorithm that comes under supervised machine learning has predicted AQI accurately compared to others. Moreover, by segregating living from non-living particulate matter and nullifying them, airborne diseases like COVID-19 can be prevented in the future.