EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture

Dinh-Dien La iD, Tien-Bang Tran, Ngoc-Huy Du iD, Ngoc-Hung Dang, Trung-Nghia Phung iD, Van-Khanh Tran iD

PLOS One · 27.07.2026 · 21 (7) · e0353166

DOI: 10.1371/journal.pone.0353166 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 27.07.2026. Verfasst von Dinh-Dien La und 5 weiteren Autor:innen. Beteiligte Einrichtung: Provincial Party Committee, Minh Xuan Ward, Tuyen Quang, Vietnam. Hauptthemen laut MEDLINE: Deep Learning, Natural Language Processing, Data Mining. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Named Entity Recognition (NER) is fundamental for automating administrative document processing in digital government systems. However, Vietnamese NLP research faces a critical infrastructure gap: existing datasets focus on generic information extraction (news, medical) rather than domain-specific administrative text. We present PAP_NER, the first large-scale, gold-standard Vietnamese administrative NER corpus comprising 162,801 sentences with 205,807 entity annotations across five entity types critical for e-Government workflows: Agency (CQ), Legal Document (VBPL), Object (ĐT), Datetime (NG), and Quantity (SL). The dataset was constructed through a rigorous human-in-the-loop annotation pipeline, achieving an inter-annotator agreement of κ = 0.85. We demonstrate PAP_NER’s value through comprehensive benchmarking of an established hybrid deep learning architecture, PhoBERT-CRF, which couples monolingual Transformer embeddings (PhoBERT) with Conditional Random Fields for structured prediction. PhoBERT-CRF achieves 97.95% Micro F1-score on the PAP_NER test set, significantly outperforming established baselines: BiLSTM+CRF (+2.01%), multilingual XLM-RoBERTa (+2.52%), and pure Transformer approaches (+0.44%). Ablation analysis reveals that the CRF layer provides statistically significant improvements for structurally complex entities (VBPL: + 0.96%, p < 0.05, McNemar’s test). We release PAP_NER publicly (DOI: 10.5281/zenodo.18044019 ) under Creative Commons BY 4.0 license to support reproducibility and enable further research in Vietnamese administrative NLP. This work establishes a foundational dataset and methodology for addressing the Vietnamese government NER gap, with implications for low-resource language NLP research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningNatural Language ProcessingData MiningVietnamHumans

Autor:innen und Institutionen

  • Provincial Party Committee, Minh Xuan Ward, Tuyen Quang, Vietnam.
  • Hanoi University of Science and Technology, Hanoi, Vietnam.
  • Institute of Artificial Intelligence, Thai Nguyen University of Information and Communication Technology, Thai Nguyen, Vietnam.
  • Faculty of Arts and Communications, Thai Nguyen University of Information and Communication Technology, Thai Nguyen, Vietnam.

ORCID-Profile: Dinh-Dien La, Ngoc-Huy Du, Trung-Nghia Phung, Van-Khanh Tran

Zitieren

APA

La, D. D., Tran, T. B., Du, N. H., Dang, N. H., Phung, T. N., & Tran, V. K. (2026). PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture. PLOS One, 21(7), e0353166. https://doi.org/10.1371/journal.pone.0353166

Vancouver

La DD, Tran TB, Du NH, Dang NH, Phung TN, Tran VK. PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture. PLOS One. 2026;21(7):e0353166. doi:10.1371/journal.pone.0353166

BibTeX
@article{La2026,
  author = {La, Dinh-Dien and Tran, Tien-Bang and Du, Ngoc-Huy and Dang, Ngoc-Hung and Phung, Trung-Nghia and Tran, Van-Khanh},
  title = {PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {7},
  pages = {e0353166},
  doi = {10.1371/journal.pone.0353166}
}
Kostenlos starten →