Air Quality Prediction and Classification using Machine Learning

Air is one of the finest resources in the present environment which contains various mixtures of gases. The various compositions may be Nitrogen, Oxygen, Sulphur Dioxide (SO2), Ozone (O3), Carbon Dioxide (CO2), Nitrogen Dioxide (NO2). The concentration of components may be varied because of air pollution. Numerous actions contribute to air pollution, which in turn makes people sick. Thus, air quality monitoring is a required task across many industrial and urban regions across the world. Thus, the researchers introduced various technologies like Internet of Things (IOT), Deep learning and Machine learning to forecast the air quality. From the state of art, the existing work considers only limited attributes which results in less accuracy in predicting air quality. As a result, this work proposes Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm along with Linear Regression which gives great classification performance when selecting the most contaminated sectors. Air Quality Operational Centers (AQOCs) are developed in places where there is severe air pollution. Experiment, which approximate the complicated physicochemical processes involved, have been the major foundation for modelling air pollution events. Even if these models have become more intricate and complicated over time, their use for real-time air quality monitoring seems unsuitable in terms of performance, input data requirements, and respect to the problem's time constraints.