Integrating Machine Learning Into Systematic Reviews: Strengthening Evidence Synthesis

• In Event: Leveraging Artificial Intelligence and Automation Tools for Systematic Reviewing

Abstract

While the application of machine learning has been rising in educational research, it has not been examined adequately in the field of research synthesis. In this paper, we describe how we use machine learning techniques to assist with a large-scale literature meta-synthesis project in undergraduate STEM education. We highlight how methodologically human screening and coding and machine learning techniques can support each other to leverage the processes of conducting a systematic review. Specifically, we describe the tools and methods we use to integrate human researchers’ qualitative endeavor with the quantitative machine learning analyses. We aim to disseminate this mixed-methods approach for widening the application of machine learning techniques in future similar systematic review projects.

Authors

• Emily Bolger, Michigan State University

• Ying Wang, FHI 360

• Rachel L. Renbarger, FHI 360

• M. Danny Caballero, Michigan State University

• Marius Nwobi, Michigan State University

• Cassandra Lem, Michigan State University

• Noah Finkelstein, University of Colorado - Boulder

• Scott P. Simkins, North Carolina A&T State University

• Taylor Boyd, Western Michigan University

• Charles R. Henderson, Western Michigan University

• Andrea L. Beach, Western Michigan University