The Rise of AI-Driven Scientometrics: A Decade of Transformative Growth (2012-2023)

Abstract

The integration of artificial intelligence (AI) and machine learning (ML) into scientometrics has revolutionized the analysis of scientific literature and the impact of the search. This study presents a comprehensive scientific analysis of the growing impact of AI and ML in the field of scientometrics from 2012 to 2023. Using large-scale bibliometric data from major scientific databases like SCOPUS, Google scholar and also collected data from crossref, we use advanced data mining techniques, natural language processing and network analysis. to map the evolving landscape of AI-driven scientometrics.

Our results reveal a significant increase in publications combining AI/ML and scientometrics, with a compound annual growth rate of over the study period. We identify five main research groups: predictive bibliometrics, automated research evaluation, scientific mapping and visualization, citation analysis and recommendation systems, and research impact assessment. The analysis also reveals changing methodological trends, with a notable departure from traditional statistical approaches towards more sophisticated AI techniques, particularly in the areas of deep learning and graph neural networks.

This study contributes to understanding how AI and ML are reshaping scientometric research and practice, providing valuable insights for researchers, policy makers and research administrators navigating rapid quantitative landscape science studies in evolution.