计算机科学 ›› 2026, Vol. 53 ›› Issue (5): 309-318.doi: 10.11896/jsjkx.250900076
唐瑞雪, 吴利琴, 钱清
TANG Ruixue, WU Liqin, QIAN Qing
摘要： 命名实体识别(NER)是自然语言处理中的核心任务之一,被广泛应用于信息抽取、问答系统、知识图谱构建等领域。然而,现有方法在处理中文文本中的嵌套实体和边界模糊问题时,仍面临多尺度特征利用不足、实体边界识别不准确等问题。为此,提出了一种面向中文命名实体识别的模型。该模型基于自适应注意力(AAM)与边界增强(BEM)机制,专门针对中文词汇无显式分隔、语义结构复杂等语言特性设计。模型通过自适应注意力机制动态融合局部与全局上下文特征,增强对中文复杂语义结构的建模能力;同时引入边界增强模块,利用深度卷积强化实体边界感知,有效缓解中文文本中嵌套实体与歧义边界带来的识别误差。实验结果表明,模型在 ACE2005-Chinese 和 Cnerta 两个中文嵌套数据集上的 F1 值分别达到 94.39% 和 83.72%,在 Weibo,Ontonotes和 Resume 这3个中文非嵌套数据集上的F1 值分别为 77.75%,84.88% 和 96.36%,均优于现有主流中文命名实体识别方法,验证了其在复杂中文文本场景下的有效性与泛化能力。
中图分类号:
| [1]QUAN Y Q,ZHANG H T,YANG B,et al.A review of named entity recognition based on deep learning[J].Microelectronics &Computer,2026,43(2):8-21. [2]PENG B,LI Y D,GONG X F,et al.A method of entity relationextraction based on heterogeneous graph neural network and text semantic enhancement[J].Computer Science,2024,51(S1):268-272. [3]WANG X,XU Y,HE X,et al.Reinforced negative sampling over knowledge graph for recommendation[C]//Proceedings of The Web Conference 2020.2020:99-109. [4]KHALID N,MOULAY A.Transformer models used for text-based question answering systems[J].Applied Intelligence:The International Journal of Artificial Intelligence,Neural Networks,and Complex Problem-Solving Technologies,2023,53(9):10602-10635. [5]LI Y H,SONG L,ZHANG C.Sparse conditional hidden Markov model for weakly supervised named entity recognition[C]//Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining(KDD’22).New York:ACM,2022:978-988. [6]PANCHENDRARAJAN R,AMARESAN A.Bidirectional LSTM-CRF for named entity recognition[C]//The 32nd Pacific Asia Conference on Language,Information and Computation(PACLIC 32).2019. [7]DUAN J Y,ZHU Y F,WANG H,et al.Chinese nested named entity recognition based on position embedding and multi-level prediction[J].Computer Engineering,2023,49(12):71-77. [8]JIA L R Z,LIU S Q,LIU Y,et al.Chinese nested named entity recognition based on hierarchical ERNIE model[J].Journal of Northeast Normal University(Natural Science Edition),2023,55(1):97-103. [9]LI X,ZHANG J W.Chinese nested named entity recognitionbased on masked SwinTransformer and boundary smoothing[J].Journal of North China University of Technology,2024,36(5):39-48. [10]WANG X H,XU Y B.Chinese named entity recognition algorithm with soft attention mask embedding[J].Journal of Jilin University(Engineering and Technology Edition),2026,56(1):231-238. [11]LU W,ROTH D.Joint mention extraction and classification with mention hypergraphs[C]//Proceedings of the 2015 Confe-rence on Empirical Methods in Natural Language Processing.2015:857-867. [12]MUIS A O,LU W.Learning to recognize discontiguous entities-supplementary[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing.2016:75-84. [13]WANG B,LU W.Neural segmental hypergraphs for overlapping mention recognition[C]//Proceedings of the 2018 Confe-rence on Empirical Methods in Natural Language Processing.ACL,2018:204-214. [14]HUANG R,CHEN Y,HUANG R.A Levitated Controlled Attention for Named Entity Recognition[J].Cognitive Computation,2025,17(1):1-15. [15]HE A K,CHEN Y P,HU Y,et al.A named entity recognition method fusing boundary interaction information[J].Journal of Guangxi Normal University(Natural Science Edition),2025,43(3):1-11. [16]HU J L,CHEN Y P,QIN Y B,et al.A Chinese named entity recognition method based on boundary mask[J/OL].Computer Engineering,1-12[2026-03-16].https://doi.org/10.19678/j.issn.1000-3428.0070482. [17]ZHU P,CHENG D,YANG F,et al.Improving Chinese named entity recognition by large-scale syntactic dependency graph[J].IEEE/ACM Transactions on Audio,Speech,and Language Processing,2022,30:979-991. [18]LIU Y,HUANG S B,LI R S,et al.USAF:Multimodal Chinese named entity recognition using synthesized acoustic features[J].Information Processing & Management,2023,60(3):103290. [19]ZHU H H,ZHANG K,LIU Y D,et al.A Chinese Named Entity Recognition Model Fusing Part-of-Speech Features[J].Computerand Digital Engineering,2025,53(6):1669-1674,1703. [20]ZHANG B H,CAI J H,ZHANG H P,et al.VisPhone:Chinese named entity recognition model enhanced by visual and phonetic features[J].Information Processing & Management,2023,60(3):103314. [21]YANG X F,FAN Y,LI Z Q,et al.Chinese Named Entity Re-cognition Model Integrating Multi-stage Features[J].Computer Engineering and Design,2025,46(1):37-43. [22]ZHANG H,QIN D H,BAI F B,et al.Chinese Named Entity Recognition Integrating Multi-level Chinese Character Features and Text Local Features[J].Journal of Chinese Information Processing,2024,38(9):93-107. [23]LIU R,GUO X,ZHU H,et al.A text-speech multimodal Chinese named entity recognition model for crop diseases and pests[J].Scientific Reports,2025,15:5429. [24]GENG R,CHEN Y,HUANG R,et al.Planarized sentence representation for nested named entity recognition[J].Information Processing & Management,2023,60:103352. [25]YAN H,SUN Y,LI X,et al.An embarrassingly easy but strongbaseline for nested named entity recognition[C]//Proceedings of the 61st Annual Meeting of the Association for ComputationalLinguistics.ACL,2023:1442-1452. [26]GUO X,CHEN Y,TANG R,et al.Camouflaged named entity recognition in 2D sentence representation[J].Expert Systems with Applications,2024,257:125096. |
|
||
首页