Evolutionary Trends in Machine Learning for Water Quality Prediction: A Visual Analysis Using CiteSpace and RStudio Bibliometrix

Machine learning, as a core algorithm of artificial intelligence, excels at discerning patterns in historical water environment data and predicting outcomes for unknown data. However, there is a lack of reports that present the current status of related studies in a comprehensive and objective manner. To assess the technological state of machine learning models used in water quality prediction, a comprehensive search was conducted for literature from 1993 to 2023 in the Web of Science Core Collection, yielding 1728 papers. This study employed the CiteSpace visualization tool to analyze the hotspots and technological advancements in machine learning for water quality prediction, providing a thorough overview of trends, including contributing countries, keyword hotspot timezone maps, and disciplines. Additionally, the Bibliometrix tool, utilizing the R language, enabled visual analysis of the field's core authors, journals, institutions, and the evolution of thematic keywords. The findings indicate a steady increase in publications over the past three decades, with technological progress marked by three distinct phases. The current research hotspots mainly focus on the optimization and improvement of models. As research hotspots continue to evolve and expand, future research integrating Internet of Things technology, satellite remote sensing, image recognition, and time‐series models for water quality prediction and assessment is expected to become a focal area for scholars.