A Bibliometric Analysis of Explainable Artificial Intelligence in Trustworthy System Development (1968–2024)
Abstract
Artificial intelligence and machine learning-based models are widely used across various domains, including healthcare, finance, cybersecurity, agriculture, education, and transportation. However, the black box nature of these AI models limits the trust in critical areas like healthcare, finance, etc. Explainable AI (XAI) emerges as an important solution to enhance transparency, interpretability and trust in the AI-based system. The study presented a bibliometric analysis of 3840 studies from the Web of Science database published between 1968 and 2024. Biblioshiny and VOSviewer software are used to show publication trends, leading countries, etc. The analysis indicates that the work is rapidly increasing in XAI-related studies after 2018, with major contributions from the USA and China. Information Fusion emerges as a leading publication source. Applications of XAI across various domains are discussed, highlighting its potential in improving the reliability and usability of AI applications in critical fields. The study helps to understand the use cases of XAI in different applications for future interdisciplinary research.
