Machine Learning in Management Research: An Integrative Typology, Review, and Research Agenda

Abstract

Machine learning research has a long history in several disciplines, but its use and applications in the management field remain limited and disjointed. Recent technological advances in computational power and architectures, increasing availability of big datasets, and innovative algorithms have transformed the implications that ML holds for management scholarship. These implications make it necessary to develop a systematic and systemic understanding of ML in management research. Towards that end, we first advance an integrative typological framework that specifies and clarifies major categories of ML for different research purposes. After clarifying how each ML category may look like in management research, we then use the framework to provide an integrative review of all published ML studies in the field. We conclude with an integrated agenda for future research informed by the framework and results of review.