Prediction of particulate matter concentration in air using data driven machine learning approach

• Main Author: Yang, Peishi
• Other Authors: Wong Kin Shun, Terence
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Machine learning; Air particulate matter
• Online Access: https://hdl.handle.net/10356/177150
• Institution: Nanyang Technological University

Air pollution is a significant issue in the world which results in many negative health impacts and significant number of deaths every year. Therefore, it is important to be able to predict future concentration of air pollutants so that the public will be able to take precautionary measures against high level of air pollution. This paper will focus on PM2.5 and PM10 pollutants in Hong Kong because these particles are smaller in size and lighter in weight. Hence, they tend to stay longer in air and will penetrate deeper into human lungs. Hong Kong is chosen as the place of interest because it is one of the most densely populated cities in the world. The aim of this paper is to employ different machine learning methods such as decision tree, support vector machine and long short-term memory network to predict the next day concentration of PM2.5 and PM10. Meteorological data including temperature, humidity and amount of rainfall will be used to provide more analysis of pollutant concentration. Data of past concentration of PM2.5 and PM10 will be downloaded from Hong Kong environmental protection department and three stations will be analyzed. Meteorological data will be downloaded from the Hong Kong Observatory website.