融合多特征的中文集成实体链接方法

doi:10.3969/j.issn.1006-2475.2019.01.013

计算机与现代化

融合多特征的中文集成实体链接方法

(河海大学计算机与信息学院,江苏南京211100)

收稿日期:2018-06-20 出版日期:2019-01-30 发布日期:2019-01-30
作者简介:冯钧(1969-),女,江苏常州人,教授,博士,CCF会员,研究方向:时空数据管理,智能数据处理与数据挖掘,水利信息化; 柳菁铧(1993-),女,硕士研究生,研究方向:信息抽取,知识图谱，E-mail: 1576236625@qq.com; 孔盛球(1990-),男,硕士研究生,研究方向:信息检索,知识图谱。
基金资助:
国家重点研发计划(2017YFC0405806); 国家自然科学基金面上项目(61602151,61370091)

Chinese Collective Entity Linking Method Based on Multiple Features

(College of Computer and Information, Hohai University, Nanjing 211100, China)

Received:2018-06-20 Online:2019-01-30 Published:2019-01-30

摘要/Abstract

摘要： 实体链接技术是将文本中的实体指称项正确链接到知识库中实体对象的过程，对知识库扩容起着关键作用。针对传统的实体链接方法主要利用上下文相似度等表层特征，而且忽略共现实体间的语义相关性，提出一种融合多特征的集成实体链接方法。首先结合同义词表、同名词表产生候选实体集，然后从多角度抽取语义特征，并将语义特征融合到构建的实体相关图中，最后对候选实体排序，选取top1实体作为链接目标。在NLP&CC2013中文微博实体链接评测数据集上进行实验，获得90.97%的准确率，与NLP&CC2013中文微博实体链接评测的最优系统相比，本文系统具有一定的优势。

关键词: 中文集成实体链接, 知识图谱, 实体消歧

Abstract: Entity linking is the process of mapping entity mentions in a document to their entities in Knowledge Base(KB) and plays a key role in the expansion of knowledge base. Aiming at traditional entity linking methods, which mainly utilize surface features such as context similarity and ignore the semantic correlation between co-occur mentions in a text corpus, a collective entity linking method based on multiple features is proposed. Firstly, it combines synonym list and namesake list to produce a set of candidate entities. After that, it extracts varieties of the semantic features and builds a referent graph. At last, it ranks the candidate entities and choses the top1 entity as the linking target. The evaluation on data sets of NLP&CC2013 Chinese micro-blog entity linking track shows a average accuracy of 90.97%, which is better than the state-of-art result.

Key words: Chinese collective entity linking, knowledge graph, entity disambiguation

中图分类号:

TP391.1

冯钧,柳菁铧,孔盛球. 融合多特征的中文集成实体链接方法[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2019.01.013.

FENG Jun, LIU Jing-hua, KONG Sheng-qiu. Chinese Collective Entity Linking Method Based on Multiple Features[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2019.01.013.

参考文献

［1］ HEATH T, MOTTA E. Ease of interaction plus ease of integration: Combining Web2.0 and the semantic Web in a reviewing site［J］. Web Semantics: Science, Services and Agents on the World Wide Web, 2008,6(1):76-83.
［2］刘峤,李扬,段宏,等. 知识图谱构建技术综述［J］. 计算机研究与发展, 2016,53(3):582-600.
［3］舒佳根,惠浩添,钱龙华,等．一个中文实体链接语料库的建设［J］．北京大学学报(自然科学版), 2015,51(2):321-328．
［4］ LI Y, WANG C, HAN F Q, et al. Mining evidences for named entity disambiguation［C］// Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2013:1070-1078.
［5］高艳红,李爱萍,段利国. 面向实体链接的多特征图模型实体消歧方法［J］. 计算机应用研究, 2017,34(10):2909-2914.
［6］邢富坤．基于维基百科的领域实体发现研究［J］．计算机应用研究, 2015,32(2):347-350,367．
［7］姚宇峰．一种新的重名消解算法在保险领域中的应用研究［J］．计算机应用研究, 2012,29(3):994-997.
［8］ GUO Y H, QIN B, LI Y Q, et al． Improving candidate generation for entity linking［C］// Proceedings of Natural Language Processing and Information Systems． 2013:225-236.
［9］ AMIT B, BRECK B. Entity-based cross-document coreferencing using the vector space model［C］// Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics. 1998:79-85.
［10］PEDERSEN T, PURANDARE A, KULKARNI A. Name discrimination by clustering similar contexts［C］ //Proceedings of the 6th International Conference on Intelligent Text Processing and Computation Linguistics. 2005:220-231.
［11］HAN X P, ZHAO J. Named entity disambiguation by leveraging wikipedia semantic knowledge［C］// Proceedings of the 18th ACM Conference on Information and Knowledge Management. 2009:215-224.
［12］BUNESCU R, PASCA M. Using encyclopedic knowledge for named entity disambiguation［C］// Proceedings of the 11th Conference of the European Chapter of the Association for Computational Linguistics(EACL06). 2006:9-16.
［13］SEN P. Collective context-aware topic models for entity disambiguation［C］// Proceedings of the 21st International Conference on World Wide Web. 2012:729-738.
［14］ SHEN W, WANG J Y, LUO P, et al. Linden: Linking named entities with knowledge base via semantic knowledge［C］// Proceedings of the 21st International Conference on World Wide Web. 2012:449-458.
［15］杨光,刘秉权,刘铭. 基于图方法的命名实体消歧［J］.智能计算机与应用, 2015(5):52-55.
［16］左乃彻. 基于中英文维基百科的命名实体消歧［D］. 北京:北京邮电大学, 2014.
［17］ZENG Y, WANG D S, ZHANG T L, et al. Linking entities in short texts based on a Chinese semantic knowledge base［C］// Proceedings of the 2nd CCF Conference on Natural Language Processing and Chinese Computing. 2013:266-276.
［18］KULKARNI S, SINGH A, RAMAKRISHNAN G, et al. Collective annotation of Wikipedia entities in Web text［C］// Proceedings of the 15th International Conference on Knowledge Discovery and Data Mining(KDD09). 2009:457-466.
［19］ZHANG W, SU J, TAN C L, et al. Entity linking leveraging automatically generated annotation［C］// Proceedings of the 23rd International Conference on Computational Linguistic(COLING10). 2010:1290-1298.
［20］王振振,何明,杜永萍. 基于LDA主题模型的文本相似度计算［J］. 计算机科学, 2013,40(12):229-232.
［21］张涛,刘康,赵军. 一种基于图模型的维基概念相似度计算方法及其实体链接系统中的应用［J］. 中文信息学报, 2015,29(2):58-67.
［22］AYMAN A, ROBERT G. Graph ranking for collective named entity disambiguation［C］// Proceedings of the 52nd Annual Meeting of the Association for Computation Linguistics(ACL14). 2014:75-80.

[1]	王运乾, 王以松, 陈攀峰, 邹龙. 融合注意力机制的药用植物文本命名实体识别[J]. 计算机与现代化, 2021, 0(11): 100-105.
[2]	王钰蓥, 王勇. 基于知识图谱和Bi-LSTM的推荐算法[J]. 计算机与现代化, 2021, 0(09): 90-98.
[3]	耿化聪, 梁宏涛, 刘国柱. 基于知识图谱与协同过滤的饮食推荐算法[J]. 计算机与现代化, 2021, 0(08): 24-29.
[4]	张艳, 杨芳, 杨蕾, 韩奎国, 李辉. 基于知识图谱的区块链技术及电力行业应用分析[J]. 计算机与现代化, 2020, 0(12): 55-60.
[5]	张紫璇,陆佳民,姜笑,冯钧. 面向水利信息资源的智能问答系统构建与应用[J]. 计算机与现代化, 2020, 0(03): 65-.
[6]	项威. 事件知识图谱构建技术与应用综述[J]. 计算机与现代化, 2020, 0(01): 10-.
[7]	冯钧，徐新，陆佳民. 水利信息知识图谱的构建与应用[J]. 计算机与现代化, 2019, 0(09): 35-.
[8]	王通,艾中良,张先国. 基于深度学习的威胁情报知识图谱构建技术[J]. 计算机与现代化, 2018, 0(12): 21-.
[9]	申存1,2,3，黄廷磊2,3，梁霄2,3. 基于多粒度特征表示的知识图谱问答[J]. 计算机与现代化, 2018, 0(09): 5-.
[10]	王良萸. 面向碳交易领域的知识图谱构建方法[J]. 计算机与现代化, 2018, 0(08): 114-.
[11]	郭琳，翟社平，高山. 知识图谱中的语义推理算法[J]. 计算机与现代化, 2017, 0(12): 71-76.

融合多特征的中文集成实体链接方法

Chinese Collective Entity Linking Method Based on Multiple Features

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

编辑推荐

Metrics

本文评价