期刊文献+

基于CNN和深层语义匹配的中文实体链接模型 被引量:6

A Chinese entity linking model based on CNN and deep structured semantic model
下载PDF
导出
摘要 实体链接是知识图谱领域的重要研究内容,现有的实体链接模型研究大多集中在对手工特征的选择上,不能很好地利用实体间的语义信息来实现更高效的实体链接效果。故提出一个基于深度语义匹配模型和卷积神经网络的实体链接模型,候选实体生成阶段采用构造同名字典,并基于上下文进行字典扩充,通过匹配来选择候选实体集。通过卷积神经网络来捕获深层语义信息,进行特征提取,并将其作为语义匹配模型的输入,通过模型训练学习选择出最佳参数,并输出语义相似度最高的候选实体作为实体链接的结果。在NLP&CC2014_ERL数据集上较Ranking SVM模型准确率提升了3.9%,达到86.7%。实验结果表明了提出的新模型性能优于当前的主流模型。 Entity Linking is an important research content in the field of Knowledge Graph.Most of the existing entity linking models focus on the selection of manual features,which cannot make good use of the semantic information between entities to achieve better efficient entity linking effect.Therefore,an improved entity linking model based on deep structured semantic model and convolutional neural network is proposed.It captures deep semantic information and extracts features through CNN,and uses them as input of the deep structured semantic model.It selects the best parameter through model training,and outputs candidate entities with the highest semantic similarity as the result of entity linking.Compared with the ranking SVM model,the proposed model improves the accuracy by 3.9%to 86.7%on the NLP&CC2014_ERL dataset.The experimental results show that the proposed model is effective and superior to the current mainstream model in entity linking tasks.
作者 吴晓崇 段跃兴 张月琴 闫雄 WU Xiao-chong;DUAN Yue-xing;ZHANG Yue-qin;YAN Xiong(College of Information and Computer,Taiyuan University of Technology,Jinzhong 030600,China)
出处 《计算机工程与科学》 CSCD 北大核心 2020年第8期1514-1520,共7页 Computer Engineering & Science
基金 国家自然科学基金(61503273)。
关键词 实体链接 知识图谱 卷积神经网络 深层语义模型 语义相似度 entity linking knowledge graph convolutional neural network deep structured semantic model semantic similarity
  • 相关文献

参考文献7

二级参考文献186

  • 1http://www.wikipedia.org/.
  • 2刘克彬,李芳,刘磊,韩颖.基于核函数中文关系自动抽取系统的实现[J].计算机研究与发展,2007,44(8):1406-1411. 被引量:59
  • 3史树明.自动和半自动知识提取[J].中国计算机学会通讯,2013.9(8):65-73.
  • 4S Auer, C Bizer,G Kobilarov, et al. Dbpedia: A Nu-cleus for Web of Open Data [ C]//Proceedings ofISWC, 2007:11-15.
  • 5Marius Pasca. Outclassing Wikipedia in Open-domainInformation Extraction: Weakly-supervised Acquisi-tion of Attributes over Conceptual Hierarchies [C]//Proceedings of the 12th Conference of the EuropeanChapter of the ACL, 2009: 639-647.
  • 6Simone Palo Ponzetto,Michael Strube. KnowledgeDerived from Wikipedia for Computing Semantic Relat-edness[J]. Journal of Artificial Inteeligence Research,2007: 181-212.
  • 7Angela Fogarolli. Word Sense Disambiguation basedon Wikipedia link structure [C]//Proceedings of Inter-national Conference on Semantic Computing, 2009 : 77-82.
  • 8P McNamee, H Simpson, H T Dang. Overview of theTAG 2009 Knowledge Base Population Track [C]//Proceedings of TAC,2009.
  • 9X Han, J Zhao. Named Entity Disambiguation by Le-veraging Wikipedia Semantic Knowledge [ C]//Pro-ceedings of CIKM,2009 : 215-224.
  • 10E Gabrilovich, S Markovitch. Computing SemanticRelatedness using Wikipedia-based Explicit SemanticAnalysis [ C]//Proceedings of IJCAI, 2007 : 1606-1611.

共引文献1430

同被引文献55

引证文献6

二级引证文献13

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部