EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A method for English paragraph grammar correction based on differential fusion of syntactic features

Weiling Liu iD, Caijun Zhao, Yongyi Li, Chenglong Cai, Hong Liu, Ruilin Qiu, Ruoci Su, Bingbing Li

PLOS One · 2025

DOI: 10.1371/journal.pone.0326081 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Weiling Liu und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of International Studies, College of Computer Science, Beibu Gulf University, Qinzhou, P. R. China. Hauptthemen laut MEDLINE: Language, Natural Language Processing, Linguistics. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: 2022 Guangxi Higher Education Undergarduate Teaching Reform Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The new progress of deep learning and natural language processing technology has strongly promoted the development of English grammar error correction. However, the existing methods mostly rely on large-scale corpus, and often ignore the fine syntactic correlation in paragraphs, which limits the efficiency in complex grammar error correction scenarios. In order to break through this bottleneck, this study proposes an innovative method to effectively use syntactic features to improve the quality and accuracy of paragraph-level grammar correction. Firstly, the sentence vector representation is constructed by BERT, and then the syntactic structure is extracted by dependency parsing. Then carry out difference fusion analysis, measure the syntactic differences of adjacent sentences by cosine similarity, identify the significant differences caused by grammatical errors according to the preset threshold, lock the position and type of errors, and input the original sentence vector into the Seq2Seq model based on Transformer. The model focuses on the wrong area by attention mechanism to generate correction suggestions. The preliminary results show that this method is significantly better than the existing grammar error correction system. In CoLA dataset, the accuracy is 0.88, which is three percentage points higher than that of BERT-GC. The accuracy of LCoLE dataset is 0.86, which is ahead of the baseline model. The accuracy of FCE data set is 0.89, which has obvious advantages. The accuracy is improved by 3% to a higher level. It shows the excellent effect of this method in grammar error recognition and correction, and has far-reaching significance in providing accurate error correction suggestions, helping English learners improve their writing ability and ensuring the quality of English writing. This study not only presents a powerful approach to English grammar error correction, but also highlights the key value of syntactic features in optimizing natural language processing applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

LanguageNatural Language ProcessingLinguisticsHumansDeep LearningSemantics

Autor:innen und Institutionen

  • College of International Studies, College of Computer Science, Beibu Gulf University, Qinzhou, P. R. China.

ORCID-Profile: Weiling Liu

Förderung

  • 2022 Guangxi Higher Education Undergarduate Teaching Reform Project

Zitieren

APA

Liu, W., Zhao, C., Li, Y., Cai, C., Liu, H., Qiu, R., Su, R., & Li, B. (2025). A method for English paragraph grammar correction based on differential fusion of syntactic features. PLOS One. https://doi.org/10.1371/journal.pone.0326081

Vancouver

Liu W, Zhao C, Li Y, Cai C, Liu H, Qiu R, et al. A method for English paragraph grammar correction based on differential fusion of syntactic features. PLOS One. 2025. doi:10.1371/journal.pone.0326081

BibTeX
@article{Liu2025,
  author = {Liu, Weiling and Zhao, Caijun and Li, Yongyi and Cai, Chenglong and Liu, Hong and Qiu, Ruilin and Su, Ruoci and Li, Bingbing},
  title = {A method for English paragraph grammar correction based on differential fusion of syntactic features},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0326081}
}
Kostenlos starten →