基于重启随机游走的实体识别与链接方法

doi:10.13190/j.jbupt.2016-220

北京邮电大学学报 ›› 2017, Vol. 40 ›› Issue (6): 115-119.doi: 10.13190/j.jbupt.2016-220

基于重启随机游走的实体识别与链接方法

谭咏梅¹, 郑迪¹, 刘姝雯¹, 吕学强²

1. 北京邮电大学智能科学与技术中心, 北京 100876;
2. 北京信息科技大学网络文化与数字传播北京市重点实验室, 北京 100101

收稿日期:2016-12-12 出版日期:2017-12-28 发布日期:2017-12-28
作者简介:谭咏梅(1975-),女,副教授,E-mail:ymtan@bupt.edu.cn.
基金资助:
网络文化与数字传播北京市重点实验室开放课题（ICDD201703）；国家自然科学基金面上项目（61671070）

Entity Discovery and Linking Approach Based on Random Walk with Restart

TAN Yong-mei¹, ZHENG Di¹, LIU Shu-wen¹, LÜ Xue-qiang²

1. Intelligence Science and Technology Center, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Beijing Key Laboratory of Internet Culture and Digital Dissemination Research, Beijing Information Science and Technology University, Beijing 100101, China

Received:2016-12-12 Online:2017-12-28 Published:2017-12-28
Supported by:

摘要/Abstract

摘要： 提出基于重启随机游走的实体识别和链接方法，在知识库部分实体构成的图结构中进行随机游走，从而获得实体和指称的分布式表示，并由此计算出相似度最高的实体作为链接实体.该方法在2015年Tri-Lingual Entity Discovery and Linking评测任务中的F值为0.665，高于其他参赛系统.实验结果表明，本方法可以有效克服特征稀缺问题，并减轻流行度差异对实验结果造成的影响.

关键词: 实体链接, 语义相似度, 随机游走

Abstract: An entity discovery and linking approach based on random walk with restart was presented. Unified semantic representation for entities and documents-the probability distribution obtained from a random walk on a subgraph of the knowledge based was adopted. According to this distributed representation, the entities that are similar with mentions as the linking results was obtained. This method achieved 0.665 F value on entity linking section of TAC 2015 TEDL task, it performs better than other participating systems. It is illustrated that the method can overcome the feature sparsity issue and is less amenable to feature sparsity bias.

Key words: entity linking, semantic relatedness, random walk

中图分类号:

TN911.22

谭咏梅, 郑迪, 刘姝雯, 吕学强. 基于重启随机游走的实体识别与链接方法[J]. 北京邮电大学学报, 2017, 40(6): 115-119.

TAN Yong-mei, ZHENG Di, LIU Shu-wen, LÜ Xue-qiang. Entity Discovery and Linking Approach Based on Random Walk with Restart[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2017, 40(6): 115-119.

参考文献

[1] Roth Dan, Heng Ji, Chang Mingwei, et al. Wikification and beyond:the challenges of entity and concept grounding[R]. ACL 2014 Gaithersburg, MD, 2014:7-18.
[2] 谭咏梅, 杨雪. 结合实体链接与实体聚类的命名实体消歧[J]. 北京邮电大学学报, 2014, 37(5):36-40. Tan Yongmei, Yang Xue. An named entity disambiguation algorithm combining entity linking and entity clustering[J]. Journal of Beijing University of Posts and Telecommunications, 2014, 37(5):36-40.
[3] 李茂林. 基于主题敏感的重启随机游走实体链接方法[J]. 北京大学学报(自然科学版), 2016, 52(1):17-24. Li Maolin. An entity linking approach based on topic-sensitive random walk with restart[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2016, 52(1):17-24.
[4] Mihalcea Rada, Csomai Andras. Wikify! linking documents to encyclopedic knowledge[C]//Sixteenth ACM Conference on Information and Knowledge Management. Lisboa, Portugal:ACM, 2007:233-242.
[5] 谭咏梅, 王睿, 李茂林. 基于上下文信息和排序学习的实体链接方法[J]. 北京邮电大学学报, 2015, 38(5):33-36. Tan Yongmei, Wang Rui, Li Maolin. An entity linking approach based on context information and learning to rank[J]. Journal of Beijing University of Posts and Telecommunications, 2015, 38(5):33-36.
[6] 怀宝兴, 宝腾飞, 祝恒书, 等. 一种基于概率主题模型的命名实体链接方法[J]. 软件学报, 2014, 25(9):2076-2087.
[7] Cucerzan Silviu. Large-scale named entity disambiguation based on wikipedia data[C]//Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning. Prague, Czech Republic:EMNLP-CoNLL, 2007:708-716.
[8] Han Xianpei, Sun Le, Zhao Jun. Collective entity linking in web text:a graph-based method[C]//International ACM SIGIR Conference on Research and Development in Information Retrieval. Beijing:ACM, 2011:765-774.
[9] Alhelbawy Ayman, Gaizauskas Robert. Graph ranking for collective named entity disambiguation[C]//Proceedings of the 52^nd Annual Meeting of the Association for Computational Linguistics. Baltimore, Maryland:ACL, 2014:75-80.
[10] Guo Zhaochen, Barbosa Denilson. Robust entity linking via random walks[C]//ACM International Conference on Information and Knowledge Management. Shanghai:ACM, 2014:499-508.
[11] Ellis J, Getman J, Fore D, et al. Overview of linguistic resources for the TAC KBP 2015 evaluations:Methodologies and results[C]//Proceedings of TAC KBP 2015 Workshop. Gaithersburg, Maryland:National Institute of Standards and Technology, 2015:16-17.

基于重启随机游走的实体识别与链接方法

Entity Discovery and Linking Approach Based on Random Walk with Restart

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	尹雨晴, 高守婉, 王小旗, 牛强. 基于随机游走的无线传感器网络节点定位方法[J]. 北京邮电大学学报, 2018, 41(2): 75-80.
[2]	谭咏梅, 李晓光, 吕学强. 融合卷积神经网络和重启随机游走的实体链接方法[J]. 北京邮电大学学报, 2018, 41(1): 65-69.
[3]	徐前方, 王嘉春, 肖波. 融合时空上下文信息的兴趣点推荐[J]. 北京邮电大学学报, 2018, 41(1): 37-42,50.
[4]	杨雨浓, 吴振宇, 朱新宁. 面向智能诊断的语义物联网知识标注与推理框架[J]. 北京邮电大学学报, 2017, 40(4): 104-110.
[5]	杨洁, 周献中. 基于语义和语法的Web服务检索方法[J]. 北京邮电大学学报, 2015, 38(6): 49-54.
[6]	谭咏梅, 王睿, 李茂林. 基于上下文信息和排序学习的实体链接方法[J]. 北京邮电大学学报, 2015, 38(5): 33-36.
[7]	谭咏梅, 杨雪. 结合实体链接与实体聚类的命名实体消歧[J]. 北京邮电大学学报, 2014, 37(5): 36-40.
[8]	王桐吴吉义王磊徐贺. WordNet中的综合概念语义相似度计算方法[J]. 北京邮电大学学报, 2013, 36(2): 98-101.
[9]	夏虹，李增智，陈彦萍. 基于概念格的语义Web服务匹配研究[J]. 北京邮电大学学报, 2006, 29(s1): 185-188.