Predicting Participation Behavior in Online Collaborative Learning through Large Language Model-Based Text Analysis

Tianjun Mo, Zihan Li and Lingfeng Guo

Pinnacle Academic Press Proceedings Series, 2025, vol. 3, 29-42

Abstract: Online collaborative learning environments generate vast amounts of textual data that contain rich behavioral patterns essential for understanding and predicting learner engagement. This research presents a comprehensive framework for predicting participation behavior in collaborative learning platforms through advanced large language model-based text analysis. Our methodology integrates sophisticated feature extraction techniques that leverage transformer-based architectures with multi-modal data fusion approaches to capture temporal behavioral dynamics. The framework employs hierarchical attention networks for behavior classification, enabling real-time identification of engagement patterns including active participation, passive engagement, collaborative leadership, and at-risk withdrawal behaviors. Experimental validation across diverse educational contexts demonstrates significant performance improvements, achieving 88.7% prediction accuracy with 156ms processing latency. The system successfully processes heterogeneous textual data from discussion forums, peer reviews, and collaborative projects across 13,267 learners from multiple institutional settings. Temporal pattern analysis reveals consistent behavioral transition patterns that enable proactive intervention strategies, with intervention success rates reaching 83.4% for collaborative conflict scenarios. The research contributes novel methodological advances in educational behavior prediction through the integration of large language models with collaborative learning analytics. The findings provide actionable insights for educational practitioners and platform designers, enabling the development of adaptive learning environments that respond dynamically to predicted behavioral changes. The framework's cross-domain generalization capabilities demonstrate practical applicability across diverse educational contexts, supporting the development of intelligent educational technologies that enhance collaborative learning outcomes through predictive behavioral analytics.

Keywords: collaborative learning; behavior prediction; large language models; text analysis (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://pinnaclepubs.com/index.php/PAPPS/article/view/172/173 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:dba:pappsa:v:3:y:2025:i::p:29-42

Access Statistics for this article

More articles in Pinnacle Academic Press Proceedings Series from Pinnacle Academic Press
Bibliographic data for series maintained by Joseph Clark ().