<![CDATA[Financial distress represents a critical stage preceding corporate bankruptcy, making its accurate prediction an urgent need for creditors, investors, and regulators. Conventional statistical models such as the Altman Z-Score (Altman, 1968) are limited in capturing non-linear relationships among financial variables, driving the rapid growth of machine learning as a methodological alternative (Ben Jabeur et al., 2023; Breiman, 2001). Although the volume of publications in this domain has risen sharply, systematic mapping of its intellectual structure, dominant actors, and thematic development remains limited. This study aims to map publication trends, country contributions, thematic cluster structures, and keyword co-occurrence networks in machine learning research for financial distress prediction, while formulating a future research agenda. Data were retrieved from Scopus, accessed on 18 July 2026, and filtered through three screening stages (open access, article document type, and exclusion of irrelevant fields), yielding a final set of 642 documents analyzed using VOSviewer (Van Eck & Waltman, 2010) with a minimum threshold of 20 keyword occurrences. The results reveal a sharp increase in publications after 2020, peaking in 2025; the dominance of China, the United States, and Taiwan; and the formation of four thematic clusters representing comparative-methodological, core-substantive, technical-optimization, and managerial-application dimensions. The novelty lies in the simultaneous integration of five bibliometric dimensions based on the most recent data up to mid-2026, accompanied by twenty-two research agendas linking thematic gaps with emerging topics such as explainable AI (Bussmann et al., 2021), generative AI (Kim et al., 2024), and graph neural networks (Li et al., 2022). Practically, this study guides academics and risk-management practitioners in directing research and adopting financial distress prediction models that are more relevant to current artificial intelligence developments.]]>
Copyrights © 2026
