Emotional Dynamic Modeling in Student Collaborative Learning Based on Natural Language Processing

DOI

10.2991/978-2-38476-551-5_50How to use a DOI?

Keywords

Natural language processing; student collaborative learning; emotional dynamic modeling; hierarchical attention; LSTM-Transformer fusion algorithm

Abstract

Aiming at the problem of emotional dynamic modeling in student collaborative learning, this study proposes a hierarchical attention LSTM-Transformer fusion algorithm (HALTF). The algorithm extracts text features through bidirectional LSTM, the hierarchical attention mechanism focuses on key emotional information, and the Transformer achieves accurate prediction. The experiment collected 5,000 sets of online collaborative learning text data, and carried out comparative experiments after preprocessing. The results show that the HALTF algorithm performs well in the emotion classification task, with an accuracy of 89.6%, an increase of 12.3% compared with the traditional LSTM algorithm; the F1 value is 88.2%, an increase of 9.5% compared with the Transformer algorithm; and the AUC reaches 0.92. The ablation experiment shows that the hierarchical attention mechanism, LSTM layer and Transformer layer are crucial to the improvement of algorithm performance. After removing the hierarchical attention mechanism, the accuracy drops to 81.2%. The algorithm effectively overcomes the problems of weak generalization ability and insufficient dynamic capture of traditional models, and provides an innovative solution for emotion analysis in the field of education.

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  - Xiaoxia Hao
PY  - 2026
DA  - 2026/03/26
TI  - Emotional Dynamic Modeling in Student Collaborative Learning Based on Natural Language Processing
BT  - Proceeding of 2025 8th International Conference on Humanities Education and Social Sciences (ICHESS 2025)
PB  - Atlantis Press
SP  - 453
EP  - 461
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-551-5_50
DO  - 10.2991/978-2-38476-551-5_50
ID  - Hao2026
ER  -