Bibliometric Analysis on Global Research Trends in Air Pollution Prediction Research Using Machine Learning from 1991–2023 Using Scopus Database

Abstract

There are a significant number of global and regional studies on air pollution prediction using machine learning. This study looks at the application of machine learning to anticipate air pollution, as well as the state of the field right now and its projected expansion. This study searches over 1794 documents created by 5354 academics and published in 745 publications between 1991 and 2023, using Scopus as the primary search engine. For the purpose of identifying and visualising major authors, journals, countries, research publications, and key trends on these concerns, articles published on these themes were evaluated using Biblioshiny, Vosviewer and S-curve analysis. We discover that interest in this subject began to grow in 2017 and has since grown at a rate of 18.56 percent per year. Although prestigious journals such as Environmental Pollution, Atmospheric Environment, and Science of the Total Environment have been at the forefront of advancing research on the application of machine learning to forecast air pollution, these journals are not the only ones doing so. The top four leading countries in terms of total citations are China (6,784 citations), the United Kingdom (2,758 citations), the United States (2145 citations), and India (1,117 citations). The top three most prestigious universities are Fudan University, China (63 articles), the University of Southern California, USA (60 articles), and Tsinghua University, China (56 articles). The authors' keyword co-occurrence network mappings show that machine learning (577 occurrences), air pollution (282 occurrences), and air quality (166 occurrences) are the top three most frequent keywords, respectively. This research focuses on using machine learning to predict air pollution.