Systematic Review of Specialized NLP Pipelines for Domain-Specific Named Entity Recognition: Architectures, Efficacy, and Research Gaps
Advances in Engineering and Information Sciences · 17.06.2026 · 2 (1) · 1-18
DOI: 10.54610/aeis.v2i1.217 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Advances in Engineering and Information Sciences, online erschienen am 17.06.2026. Verfasst von Jason Sigfred Seril. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
A systematic thematic review of 261 publications (2020–2025) reveals the current landscape of domain-adapted Named Entity Recognition pipelines, with biomedical/clinical domains dominating (37.9% of works) and emerging financial/agricultural applications identified as underrepresented. The customization of encoder-based architectures (26.4%), especially domain-pretrained variants such as BioBERT and ClinicalBERT, stands out as the prevalent approach, complemented by knowledge/ontology integration (10.3%) and emerging use of LLM components (14.2%). Transfer learning (23.0%) and fine-tuning (15.3%) are identified as the most common techniques for specialization, while prompt engineering (5.4%) and few-shot learning (3.4%) are noted as emerging annotation-efficient strategies. Domain-specific terminology handling is reported as the most common benefit (41.8%), while the main reported open challenge is the ambiguity of entities (6.9%), followed by domain transfer limitations (6.5%) and scarcity of resources for low-resource languages (5.7%). The systematic review findings position domain-specialized NER from an exploratory phase to a consolidation phase, with transfer learning, fine-tuning, and knowledge integration becoming the prevailing paradigm, while ongoing challenges in semantic understanding and domain transferability present opportunities for future work on annotation-efficient methods, the formalization of domain characterization, and improving research coverage of underexplored application domains.
Zitieren
APA
Seril, J. S. (2026). Systematic Review of Specialized NLP Pipelines for Domain-Specific Named Entity Recognition: Architectures, Efficacy, and Research Gaps. Advances in Engineering and Information Sciences, 2(1), 1-18. https://doi.org/10.54610/aeis.v2i1.217
Vancouver
Seril JS. Systematic Review of Specialized NLP Pipelines for Domain-Specific Named Entity Recognition: Architectures, Efficacy, and Research Gaps. Advances in Engineering and Information Sciences. 2026;2(1):1-18. doi:10.54610/aeis.v2i1.217
BibTeX
@article{Seril2026,
author = {Seril, Jason Sigfred},
title = {Systematic Review of Specialized NLP Pipelines for Domain-Specific Named Entity Recognition: Architectures, Efficacy, and Research Gaps},
journal = {Advances in Engineering and Information Sciences},
year = {2026},
volume = {2},
number = {1},
pages = {1--18},
doi = {10.54610/aeis.v2i1.217}
}
EUVIMED