Machine Learning Measure-Supported Empirical Studies: A Survey

Machine learning and data science techniques have been widely used in practice and have received significant attention. In research settings, business scholars have begun to adopt machine learning as methodological tools. This paper reviews the literature that leverages machine learning-generated measurements to support empirical analysis and yield business insights, where machine learning can reduce reliance on human coding or enable new theoretical perspectives. We summarize the workflow, constructs, and measure structures in these machine learning measure-supported empirical studies. We believe machine learning holds substantial potential to advance business research.

History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Learning.

Funding: This work was supported by the Shanghai Pujiang Program, Shanghai Municipal Human Resources and Social Security Bureau [Grant 23PJC097]; Research Grants Council, University Grants Committee [Grants 11500625 and 11501722]; and National Natural Science Foundation of China [Grants 72301067 and 72502212].

Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijoc.2024.0918.