Prediction of Particulate Matter-2.5 in Coastal Region of India Using ModGA-XGB Model

The air is getting polluted due to the presence of harmful substances such as Particulate Matter(PM<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2.5</inf>, PM<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">10</inf>), Nitrogen Oxide (NO), Nitrogen Dioxide (NO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), Sulphur Dioxide (SO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), Ozone (O<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf>), Ammonia (NH<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf>), Carbon Monoxide (CO) and Volatile Organic Compounds (VOC) etc. Polluted air can cause changes in climate as well as pose several health issues in the human body. The primary goal of this study is to forecast the concentration of PM<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2.5</inf> pollutants in the air of two Indian coastal cities such as Chennai and Mumbai. The air quality data of these two mentioned cities are collected from the official website of Central Pollution Control Board, India and then the collected data are gone through several preprocessing approaches for better training of the model. The work presents a Modified Genetic Algorithm-based Extreme Gradient Boosting (ModGA-XGB) model for the prediction of PM<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2.5</inf> and the proposed model also outperformed other traditional machine learning models. The outcome of the work is highly beneficial to enhance the air quality monitoring strategy.