HREB-CRF: Hierarchical Reduced-bias EMA for Chinese Named Entity Recognition
Incorrect boundary division, complex semantic representation, and differences in pronunciation and meaning often lead to errors in Chinese Named Entity Recognition(CNER). To address these issues, this paper proposes HREB-CRF framework: Hierarchical Reduced-bias EMA with CRF. The proposed method amplifies word boundaries and pools long text gradients through exponentially fixed-bias weighted average of local and global hierarchical attention. Experimental results on the MSRA, Resume, and Weibo datasets show excellent in F1, outperforming the baseline model by 1.1\%, 1.6\%, and 9.8\%. The significant improvement in F1 shows evidences of strong effectiveness and robustness of approach in CNER tasks.
Code (1)
Tasks
Chinese Named Entity Recognitionnamed-entity-recognitionNamed Entity RecognitionMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Chinese Named Entity Recognition via Adaptive Multi-pass Memory Network with Hierarchical Tagging Mechanism
Named entity recognition (NER) aims to identify text spans that mention named entities and classify them into pre-defined categories. For Chinese NER task, most of the existing methods are character-based sequence labeli…
Chinese Named Entity Recognitionnamed-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)+2Multiple-Debias: A Full-process Debiasing Method for Multilingual Pre-trained Language Models
Multilingual Pre-trained Language Models (MPLMs) have become essential tools for natural language processing. However, they often exhibit biases related to sensitive attributes such as gender, race, and religion. In this…
parameter-efficient fine-tuningData AugmentationPromoting the linguistic diversity of TEI in the Maghreb and the Arab region
The project targets both oral corpus and the rich text resources written in the Maghreb region. It focuses particularly on the continuity, for more than 12 centuries, of a classical still alive Arabic language and on the…
DiversityBigCilin: An Automatic Chinese Open-domain Knowledge Graph with Fine-grained Hypernym-Hyponym Relations
This paper presents BigCilin, the first Chinese open-domain knowledge graph with fine-grained hypernym-hyponym re-lations which are extracted automatically from multiple sources for Chinese named entities. With the fine-…
Hrebs and Cohesion Chains as similar tools for semantic text properties research
In this study it is proven that the Hrebs used in Denotation analysis of texts and Cohesion Chains (defined as a fusion between Lexical Chains and Coreference Chains) represent similar linguistic tools. This result gives…