Abstract
Real-world named entity recognition (NER) datasets are notorious for their noisy nature, attributed to annotation errors, inconsistencies, and subjective interpretations. Such noises present a substantial challenge for traditional supervised learning methods. In this paper, we present a new and unified approach to tackle annotation noises for NER. Our method considers NER as a constituency tree parsing problem, utilizing a tree-structured Conditional Random Fields (CRFs) with uncertainty evaluation for integration. Through extensive experiments conducted on four real-world datasets, we demonstrate the effectiveness of our model in addressing both partial and incorrect annotation errors. Remarkably, our model exhibits superb performance even in extreme scenarios with 90% annotation noise.- Anthology ID:
- 2023.emnlp-main.872
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 14112–14123
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.872
- DOI:
- 10.18653/v1/2023.emnlp-main.872
- Cite (ACL):
- Jian Liu, Weichang Liu, Yufeng Chen, Jinan Xu, and Zhe Zhao. 2023. Addressing NER Annotation Noises with Uncertainty-Guided Tree-Structured CRFs. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 14112–14123, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Addressing NER Annotation Noises with Uncertainty-Guided Tree-Structured CRFs (Liu et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/ingest-bitext-workshop/2023.emnlp-main.872.pdf