Machine learning in accounting and finance research: a literature review

Abstract

In recent years, scholars in accounting and finance have shown a growing interest in employing machine learning for academic research. This study combines bibliographic coupling and literature review to analyze 575 papers from 93 well-established journals in the field of accounting and finance published between 1996 and 2022, and addresses three interrelated research questions (RQs): RQ1 How is research on the impact of machine learning on accounting and finance developed? RQ2 What is the focus within this corpus of literature? RQ3 What are the future avenues of machine learning in accounting and finance research? We adopt a critical approach to the research foci identified in the literature corpus. Our findings reveal an increased interest in this field since 2015, with the majority of studies focused either on the US market or on a global scale, with a significant increase in publications related to Asian markets during 2020–2022 compared to other regions. We also identify that supervised models are the most frequently applied, in contrast to unsupervised models, which mainly focus on clustering applications or topic extraction through the LDA algorithm, and reinforcement models, which are rarely applied, yield mixed results. Additionally, our bibliographic analysis reveals six clusters, and we discuss key topics, current challenges and opportunities. Finally, we outline machine learning constraints, highlighting common pitfalls, and proposing effective strategies to overcome current barriers and further advance research on this issue.