Machine Learning for Detecting Financial Statement Fraud: A Bibliometric Analysis

This study presents a bibliometric analysis of publishing, citation, and collaboration trends in financial statement fraud detection using Machine Learning (ML) from 2010 to 2023. By utilizing data from Scopus and Google Scholar, the analysis was conducted in R Studio, identifying key thematic and collaborative patterns across ML, artificial intelligence (AI), deep learning, and predictive modeling. The results highlight significant advancements in fraud detection, driven largely by international collaborations between countries such as China-Australia, Canada-UAE, and Belgium-Netherlands. Notably, fraud-related research has begun to surpass ML and AI in keyword trends, indicating a shift in focus within the field. Node analysis further exposes key topics within collaborative networks, emphasizing the growing relevance of machine learning in fraud detection. These findings offer critical insights for researchers, policymakers, and industry professionals, providing a clearer understanding of the evolving global research landscape and its implications for the future of fraud detection technologies.