Use transformer-based topic modeling (BERTopic) to map thematic structure and evolution of an emerging research field¶

Strategy · All strategies
Evidence · 1 claim (1 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description¶

The article demonstrates using BERTopic — "a transformer-based topic modeling approach that leverages contextual embeddings" — on a publication corpus to identify major research topics, track their evolution over time, and recognize future research directions. Researchers studying emerging fields can apply this pipeline to obtain a data-driven thematic overview, as shown with 1,048 AIOL publications.

Design Implications¶

Context¶

Requirements¶

• A publication corpus on the target field
• A contextual-embedding-based topic modeling approach such as BERTopic

Constraints¶

• Demonstrated on the AIOL field

Target Learners¶

• education researchers

Target Learning Goals¶

• identifying research themes and trends in a field

Claims¶

• BERTopic analysis of 1,048 AIOL publications identifies research themes including adaptive learning systems, sentiment analysis, and predictive analytics [+W]

Related Strategies¶

• Educational Research Map Visualization

Examples¶

-

Key Sources¶

• Chen, X., Xie, H., Peng, X., Tao, X., Li, L., Qin, S. J., & Wang, F. L. (2026). Artificial Intelligence in Online Learning: Using BERTopic to Track Research Topics and Their Evolutions. The International Review of Research in Open and Distributed Learning, 27(3), 67–94. https://doi.org/10.19173/irrodl.v27i3.9516