A Bibliometric and Network Analysis of Statistical Distributions in Machine Learning

Statistical distributions constitute a fundamental component of machine learning by providing the probabilistic framework required for modelling uncertainty, variability, and data-generating processes. They underpin a wide range of methodologies, including Bayesian inference, probabilistic graphical models, generative learning and uncertainty-aware artificial intelligence. Despite their widespread application, the evolution and intellectual structure of research on statistical distributions in machine learning have not been systematically examined. This study presents a bibliometric and network analysis of 840 Scopus-indexed publications published between 1997 and 2025. Bibliographic data were analysed using Biblioshiny and VOS viewer to evaluate publication trends, influential sources, author productivity, collaboration patterns and thematic developments. Descriptive indicators were complemented by co-authorship, co-citation, keyword co-occurrence and thematic analyses to explore the social and intellectual structure of the field. The findings reveal sustained growth in research output, increasing international collaboration and the dominance of a core set of journals and authors. Network analyses identify probabilistic modelling, Bayesian learning, uncertainty quantification and generative approaches as the principal intellectual foundations of the field. The results further indicate a gradual shift from classical statistical learning frameworks toward uncertainty-aware and explainable machine learning paradigms. By mapping the evolution, knowledge structure and emerging directions of this research domain, the study provides a comprehensive reference for future investigations at the intersection of statistical distribution theory and machine learning.