Mapping the intellectual structure of book recommendation systems: A scientometric analysis

The Tacit knowledge of the individuals brings transformational changes in the society when it becomes explicit knowledge. The books are the best sources of explicit knowledge even in the age of Artificial Intelligence (AI). Due to proliferation of information, the right book for the right person can be well identified when it is recommended by the who have been benefitted. This study reflects a detailed Scientometric analysis of global research works on AI-based book recommendation systems indexed in the Scopus database published between 2016 and February 2026. The analysis is done using tools Biblioshiny and VOS viewer to detail the intellectual structure, emerging areas of study, collaboration networks, and thematic evolution of the research domain. The results find out that the period between 2016 and February 2026 shows significant increase in the research works spanning around 819 publications across 473 sources, showcasing an annual growth rate of 13.87%. The leading contributors to the research works include Wang Y, Zhang Y, Wang X, Di Noia T, Anelli V.W, and Li X. Country-level analysis provides a strong contribution from China and India, followed by the United States and European countries. The thematic map representation reveals prominence of recommender systems, artificial intelligence, and natural language processing systems forming the core structure of the research field. The keyword co-occurrence network convey that central themes revolve around recommendation systems, machine learning, and deep learning while areas like natural language processing, knowledge graphs, graph neural networks, and collaborative filtering reflect the field's growing demand for more practical, robust and intelligent data-driven recommendation frameworks. The study is significant as it showcases remodelling of existing traditional model towards more robust practical application of book recommendation system. While the methodologies, systemic changes and suggestive measures often accompany with its own challenges in a real-world scenario, the paper conveys valuable insights for researchers, librarians, and system developers for their area of functioning. At the same time, study highlights need for setting up an intelligent model of AI book recommendation systems.