Research on paragraph-level functional structure recognition in scientific literature: a data augmentation method based on LLMs and lexical function

Haotan Liu (), Zhuo Chen (), Qunzhe Ding (), Jiafeng Zhang (), Jiawei Liu (), Jiming Hu () and Wei Lu ()
Additional contact information
Haotan Liu: Wuhan University, School of Information Management
Zhuo Chen: Wuhan University, School of Information Management
Qunzhe Ding: Wuhan University, School of Information Management
Jiafeng Zhang: Wuhan University, School of Information Management
Jiawei Liu: Wuhan University, School of Information Management
Jiming Hu: Wuhan University, School of Information Management
Wei Lu: Wuhan University, School of Information Management

Scientometrics, 2025, vol. 130, issue 10, No 9, 5473-5502

Abstract: Abstract The automatic recognition of functional structures within scientific literature enhances fine-grained information retrieval and mitigates issues associated with imbalanced text classification. Although multilevel functional structure research is relatively advanced, achieving high accuracy in overall label prediction and recognition, precise functional structure identification at the paragraph level remains challenging. To address this issue, we propose an innovative method called SLSG, which stands for (Synonym replacement + Lexical function based LLM Auto-labeling + SciBERT-GCN). This method serves as a data augmentation (DA) strategy for the identification of paragraph-level functional structure. Specifically, SLSG integrates several mechanisms, including synonym replacement and lexical function-based auto-annotation using Large Language Models (LLMs) for DA. It combines augmented data with a SciBERT-GCN model to effectively extract features by leveraging contextual sequence information between paragraphs. Applied to the ScienceDirect dataset, SLSG achieves an F1 score of 86% for paragraph-level functional structure recognition, marking an 18% improvement over the baseline models and demonstrating a significant enhancement in classification performance. Moreover, SLSG employs graph neural networks to capture both dependency relationships and topological structures among word nodes. This approach not only augments the representation of scientific literature, but establishes a solid research paradigm to address the challenges related to unbalanced text classification.

Keywords: Functional structure; Text generation; Data augmentation; Scientific literature; Large language Model; Lexical function (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11192-025-05355-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:spr:scient:v:130:y:2025:i:10:d:10.1007_s11192-025-05355-6

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11192

DOI: 10.1007/s11192-025-05355-6

Access Statistics for this article

Scientometrics is currently edited by Wolfgang Glänzel

More articles in Scientometrics from Springer, Akadémiai Kiadó
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().