现代信息科技2025，Vol.9Issue(16)：50-56,7.DOI:10.19850/j.cnki.2096-4706.2025.16.010
基于数据增强与多特征组合的中文命名实体识别
Chinese Named Entity Recognition Based on Data Augmentation and Multi-feature Combination
摘要
Abstract
Named Entity Recognition(NER)is an important and fundamental task in the fields of Information Retrieval and Natural Language Processing.At present,the mainstream methods based on character combination Attention Mechanism(AM)and character and word combination AM are faced with problems such as corpus,Chinese word segmentation and overflow words.Therefore,from the perspective of dataset and the combination of character and word,this paper proposes a method combining Data Augmentation(DA)and dynamic feature combination of character and word domain information.The use of DA technology improves the quality and expands the scale of corpus,while the dynamic character and word features combined with domain information by using AM provide effective textual semantic information.The paper conducts a large number of experiments on CCKS2017 and Common datasets,and the experimental results show the effectiveness of the proposed model.关键词
数据增强/动态特征组合/注意力机制/中文命名实体识别Key words
data augmentation/dynamic feature combination/Attention Mechanism/Chinese Named Entity Recognition分类
信息技术与安全科学引用本文复制引用
李源..基于数据增强与多特征组合的中文命名实体识别[J].现代信息科技,2025,9(16):50-56,7.基金项目
信阳农林学院青年教师科研基金(QN2023014) （QN2023014）
信阳农林学院科研促教学项目(kj-2022007) （kj-2022007）
