Sustainable Biofouling Control: A Bibliometric Evaluation of Machine Learning Trends in Natural Marine Antifouling Coating

Abstract

Global marine biofouling prevention and control is accelerating its transition toward sustainability. Natural product antifoulants and bionic surfaces have become the mainstream alternatives to traditional biocidal coatings. In recent years, relevant scientific research outputs in this field have grown rapidly, creating an urgent need to systematically sort out applications of machine learning in this green chemistry field, to support accelerated material development and performance prediction. This paper conducts a bibliometric evaluation of global research on machine learning applications in this field spanning 2010–2026. We retrieved 83 relevant publications from the Scopus database via a multi-layer Boolean search, adopted citation metrics, co-authorship network analysis, and keyword co-occurrence analysis methods, and used bibliometric tools for the R programming language to build visual networks and analyze the field’s internal structure. This study finds that the number of publications in the field is growing at an accelerating pace, and the research paradigm has shifted from empirical modelling to advanced algorithms such as deep learning. Three core thematic clusters have formed: machine learning-driven natural compound screening, bionic surface texture optimization, and ecotoxicity and durability prediction. Meanwhile, we identified the leading countries and journals in the field and found that cross-disciplinary international collaboration is extremely limited. This study maps the full landscape of the field, points out the gap caused by the lack of standardized datasets, and proposes a strategic roadmap to guide future research and policy formulation.