Bibliometric analysis is a vital methodological tool for mapping scientific knowledge and identifying research trends. However, challenges in structuring, sorting, and interpreting bibliometric results, as well as a lack of standardized workflows, often hinder their practical application, especially for novice researchers. This study introduces a structured methodological framework that improves the design and interpretability of bibliometric analyses. We propose and discuss four key methodological pillars: (i) formulation of precise research questions to guide design and metrics; (ii) the application of artificial intelligence (AI)-based tools for efficient and unbiased article screening; (iii) the normalization of research output data using economic indicators, such as gross domestic product, to improve cross-country comparisons and (iv) the use of dual-tool analysis—specifically VOSviewer and Biblioshiny—to validate bibliometric findings. This integrated framework improves the transparency, clarity and robustness of bibliometric studies. AI-powered article screening leads to faster and more consistent inclusion decisions. Normalizing bibliometric data based on economic output enhances the interpretability of national research performance. Cross-referencing bibliometric models with two complementary tools allows for a more robust depiction of knowledge structures and emerging themes. In conclusion, the introduction of this methodological framework advances the practice of bibliometric analysis, contributing to a more reliable and informative bibliometric approach.
Stefanis et al. (Sat,) studied this question.