Evolution of simulation scholarship: A text mining exploration

Background: Text mining uses advanced machine learning algorithms, natural language processing, and statistical analyses to unveil hidden themes in a body of text. Reviewing the simulation literature though text mining allows researchers to categorize extensive collections of publications and develop salient questions based on mapping the evolution of simulation scholarship. Methods: This review examined manuscripts in five healthcare simulation journals between 2006 and 2022, resulting in 2,382 articles included in the text corpus. Results: The top 20 topics were identified and named, in addition to which topics had the highest number of publications. Finally, publication patterns for each topic were examined, with several hypotheses offered as explanation of the results. Discussion: Practical implications of text mining include tracking publication shifts over time, as well as identifying areas of future research that warrant more in-depth, contextual analyses.