A Bibliometric Analysis of Explainable Artificial Intelligence (XAI): Trends, Themes, and Global Research Dynamics

The active development of artificial intelligence resulted in the extensive introduction of complicated designs, most of them being black boxes. This has raised an urgent requirement of Explainable Artificial Intelligence (XAI) to enhance the explainability, trustworthiness, and interpretability of decision-making systems. The purpose of this research is to evaluate the current worldwide trends and patterns of research and development of XAI through a bibliometric approach. Major scientific databases with publications were searched, and this data was analyzed with the help of bibliometric tools and visualization techniques. The findings indicate that post 2019, there are intense growths in research production which point to the increasing interest in the topic. In terms of key words, artificial intelligence, machine learning and interpretability are prominent subjects of research. Analysis on a country level shows that the United States, the United Kingdom, and Germany contributed the most. The results also reveal the emergence of cross-functional fields of research, which integrate XAI with other fields that include psychology, medicine, and data science. According to the study, XAI research is slowly increasing in a transition towards the theoretical basis to the practical and application-based methods of research. These ideas allow getting a clear vision of the present day of research in explainable artificial intelligence and its further evolution.