Artificial intelligence and machine learning in finance: Identifying foundations, themes, and research clusters from bibliometric analysis

Abstract Artificial intelligence (AI) and machine learning (ML) are two related technologies that are emergent in financial scholarship. However, no review, to date, has offered a wholistic retrospection of this research. To address this gap, we provide an overview of AI and ML research in finance. Using both co-citation and bibliometric-coupling analyses, we infer the thematic structure of AI and ML research in finance for 1986–April 2021. By uncovering nine (co-citation) and eight (bibliometric coupling) specific clusters of finance that apply AI and ML, we further identify three overarching groups of finance scholarship that are roughly equivalent for both forms of analysis: (1) portfolio construction, valuation, and investor behavior; (2) financial fraud and distress; and (3) sentiment inference, forecasting, and planning. Additionally, using co-occurrence and confluence analyses, we highlight trends and research directions regarding AI and ML in finance research. Our results provide assessment of  AI and ML in finance research.