Identifying interdisciplinary emergence in the science of science: combination of network analysis and BERTopic

This study proposes a novel approach to identify emerging interdisciplinary science using a combination of network analysis and BERTopic, an embedded topic modeling technique. The researchers constructed a global map of scientific co-occurrence networks from Web of Science data, defining interdisciplinary fields as those involving multiple science categories. By comparing co-occurrence networks across time periods, they identified changing influence patterns and used BERTopic to associate these with interdisciplinary classifications. The results demonstrate the emergence of global interdisciplinary sciences, with qualitative validation used to identify the sources of these emergent areas. The study discusses potential applications for identifying emergence through the merging of global interdisciplinary domains.

Keungoui Kim, Dieter F. Kogler, Sira Maliphol