Systematic Review of ASEE Conference Proceedings (2007-2016) with A Machine Learning Approach

摘要

This study explores the thematic structure of a large body of scholarly proceedings in Engineering Education. To this end, we first provide a mixed method that combines topic modeling, a machine learning methodology designed to extract the thematic structure of large data, with qualitative analysis of its results. Second, we identify the major topics among the engineering education studies, their trends over time, and the semantic similarities between topics that have similar trending patterns. Our dataset includes over 14,000 conference proceedings published between 2007 and 2016. Our analysis identified 26 topics that have been the focus of engineering education scholarly work, as approximated by the conference proceedings published in the American Society of Engineering Education. We report our results by providing insights on trending topics and their relationships.