Twenty-Five Years of Research on Artificial Intelligence-Driven Farmers' Decision-Making: A Bibliometric and Science Mapping Analysis of Intellectual Structure, Thematic Evolution, and Future Research Directions

<ns5:p>Background Artificial intelligence (AI) is transforming agricultural decision-making through data-driven approaches to resource management, production planning, and risk mitigation. As digital agriculture continues to expand, research on AI-supported farmers’ decision-making has increased across multiple disciplines. However, the existing body of knowledge remains fragmented, limiting a comprehensive understanding of its intellectual foundations, thematic structure, and emerging research directions. This study systematically maps the scientific landscape of AI-driven farmers’ decision-making and identifies its key contributors, thematic domains, and future research priorities. Methods A bibliometric and science-mapping approach was employed using the Scopus database. Following the PRISMA protocol, 217 English-language articles and review papers published between 2000 and 2025 were selected using the search query: “Farmers” AND “Decision-Making” AND “Artificial Intelligence”. Performance analysis and thematic mapping were conducted to examine publication trends, influential contributors, geographical distribution, conceptual structures, and thematic evolution. Results The findings reveal a rapidly expanding field, with an annual growth rate of 19.37%. India emerged as the most productive contributor, while Spain and Germany demonstrated the highest citation impact. Four principal thematic domains were identified: AI-enabled precision agriculture and decision support; smart resource monitoring and water management; environmental modelling and resource governance; and sustainable agricultural systems. Thematic evolution indicates a shift from environmental simulation and resource optimisation towards data-intensive agricultural systems supported by machine learning, the Internet of Things (IoT), forecasting, crop-yield prediction, and explainable artificial intelligence. The increasing prominence of explainable AI reflects growing attention to transparency, interpretability, and user-centred design.</ns5:p>