What do the differences and commonalities in doctoral dissertation acknowledgments across disciplines reveal?

Acknowledgments in academic dissertations occupy a unique role within scholarly communication. Prior research has investigated acknowledgments through lenses such as funding attribution, genre analysis, and linguistic features. This study examines acknowledgments in doctoral dissertations from Chinese universities, organized by broad disciplinary categories. Utilizing BERTopic modeling, the research identifies topic keywords embedded within dissertation acknowledgments. Furthermore, computational linguistics techniques are employed to quantitatively evaluate the content and stylistic attributes of these acknowledgments, complemented by hierarchical clustering analysis to explore cross-disciplinary similarities. The topic modeling results indicate that acknowledgments by Chinese doctoral students frequently convey emotional reflections and exhibit distinct disciplinary traits. Additionally, hierarchical clustering shows that disciplines with similar characteristics exhibit greater similarity in the content and writing style of their acknowledgments, indicating that academic training influences researchers' writing to some degree. This study seeks to catalyze further scholarly inquiry into this domain, advocating for expanded investigations from perspectives including psychology, neuroscience, and cross-cultural studies.