Past, present and future of the applications of machine learning in soil science and hydrology

• DOI: 10.17221/94/2022-swr
• Publication Date: 2023-03-22T08:34:05Z
• Authors: Xiangwei Wang; Yizhe Yang; Jianglong Lv; Hailong He
• Keywords: scientometric analysis; machine learning; S; 0208 environmental biotechnology; Agriculture; 02 engineering and technology; science mapping; soil

Machine learning can handle an ever-increasing amount of data with the ability to learn models from data. It has been widely used in a variety disciplines and is gaining increasingly more attention nowadays. As it challenging map soil hydrological information that are characterised high spatial temporal variability, applications machine science hydrology (AMLSH) have become popularised. To better understand current state AMLSH research, scientific quantitative approach was performed statistically analyse publication 1973 2021 archived Scopus database using scientometric analysis tools, including VOSviewer, CiteSpace, open-source R package "bibliometrix". The results show significant increase number publications on since 2006. major contributions were identified based country origins (China, USA, India), institutions (Hohai University, Islamic Azad Wuhan University), journals (Journal Hydrology, Remote Sensing, Geoderma). keywords research demonstrates four hotspots: neural network, artificial intelligence, learning, soil. most frequently utilised (ML) methods networks, decision trees, random forests other for image processing predictive analysis. McBratney et al. 2003 highly cited article. Our sheds light process concludes future perspectives.