SciClusterNet: Discovering Emerging Topics in LLM and AI in Education

DOI

10.2991/978-94-6239-664-7_76How to use a DOI?

Keywords

Topic Modeling; Trend Analysis; SciBERT; BERTopic

Abstract

The increase in research studies focused on Large Language Models (LLMs) and Artificial Intelligence in Education (AIED) has increased the difficulty of discovering emerging themes and research trajectories. This research introduces SciClusterNet, an unsupervised methodology that combines SciBERT embeddings, UMAP reduction, multiple algorithms, and BERTopic to investigate 4,000 research abstracts from the domains of LLM and AIED research. K-Means provided the highest cluster architecture (Silhouette = 0.3877, DBI = 0.7452), and BERTopic produced six coherent topics across the domains of cybersecurity, multimodal reasoning, finance, code generation, and Q&As in education. Sci-ClusterNet had a topic coherence (Cv = 0.5055, NPMI = 0.0383) that is somewhat lower than NMF (0.5784, 0.0696), yet it detects themes with significantly more (7.6% higher Silhouette, 19.2% lower DBI, and 239% CH improvement under DBSCAN) semantic richness and clustering quality than the baselines. Although the difference in topic coherence is somewhat negligible, we confirm that the SciBERT embedding generates more coherent and scientifically faithful topics than bag-of-words models. Overall, using SciClusterNet is a robust and domain-specific approach to identify emerging topics and themes in increasingly dynamic research in LLM and AI in Education research.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. S. Zobaer Ahmed
AU  - Md. Towsif Billah
AU  - Emon Safayet Rid
AU  - Md. Rehab Ansary Yasin
AU  - Tohedul Islam
PY  - 2026
DA  - 2026/06/08
TI  - SciClusterNet: Discovering Emerging Topics in LLM and AI in Education
BT  - Proceedings of the International Conference on Intelligent Data Analysis and Applications (IDAA 2025)
PB  - Atlantis Press
SP  - 1110
EP  - 1126
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-664-7_76
DO  - 10.2991/978-94-6239-664-7_76
ID  - Ahmed2026
ER  -