计算机科学技术学报 ›› 2020,Vol. 35 ›› Issue (4): 724-738.doi: 10.1007/s11390-020-0280-1

所属专题： Data Management and Data Mining

• • 上一篇    下一篇

基于搜索引擎丰富上下文信息的实体链接方法

Yi-Ting Wang1, Jie Shen1, Zhi-Xu Li1,2,3,*, Member, CCF, IEEE, Qiang Yang4, An Liu1, Member, CCF, IEEE, Peng-Peng Zhao1, Member, CCF, IEEE, Jia-Jie Xu1, Member, CCF, IEEE Lei Zhao1, Member, CCF, IEEE, Xun-Jie Yang5   

  1. 1 Institute of Artificial Intelligence, School of Computer Science and Technology, Soochow University Suzhou 215006, China;
    2 iFLYTEK Research, Suzhou 215000, China;
    3 State Key Laboratory of Cognitive Intelligence, iFLYTEK, Hefei 230000, China;
    4 Computer, Electrical and Mathematical Sciences and Engineering, King Abdullah University of Science and Technology Jeddah 23955, Saudi Arabia;
    5 Anhui Toycloud Technology Co., Ltd., Hefei 230000, China
  • 收稿日期:2020-02-08 修回日期:2020-05-20 出版日期:2020-07-20 发布日期:2020-07-20
  • 通讯作者: Zhi-Xu Li E-mail:zhixuli@suda.edu.cn
  • 作者简介:Yi-Ting Wang is a Master student in computer science and technology at Soochow University, Suzhou. Her research interest is entity linking.
  • 基金资助:
    This work was supported by the National Key Research and Development Program of China under Grant No. 2018AAA0101900, the Natural Science Foundation of Jiangsu Province of China under Grant No. BK2019 1420, the National Natural Science Foundation of China under Grant No. 61632016, the Natural Science Research Project of Jiangsu Higher Education Institution under Grant No. 17KJA5 20003, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the Suda-Toycloud Data Intelligence Joint Laboratory.

Enriching Context Information for Entity Linking with Web Data

Yi-Ting Wang1, Jie Shen1, Zhi-Xu Li1,2,3,*, Member, CCF, IEEE, Qiang Yang4, An Liu1, Member, CCF, IEEE, Peng-Peng Zhao1, Member, CCF, IEEE, Jia-Jie Xu1, Member, CCF, IEEE Lei Zhao1, Member, CCF, IEEE, Xun-Jie Yang5        

  1. 1 Institute of Artificial Intelligence, School of Computer Science and Technology, Soochow University Suzhou 215006, China;
    2 iFLYTEK Research, Suzhou 215000, China;
    3 State Key Laboratory of Cognitive Intelligence, iFLYTEK, Hefei 230000, China;
    4 Computer, Electrical and Mathematical Sciences and Engineering, King Abdullah University of Science and Technology Jeddah 23955, Saudi Arabia;
    5 Anhui Toycloud Technology Co., Ltd., Hefei 230000, China
  • Received:2020-02-08 Revised:2020-05-20 Online:2020-07-20 Published:2020-07-20
  • Contact: Zhi-Xu Li E-mail:zhixuli@suda.edu.cn
  • About author:Yi-Ting Wang is a Master student in computer science and technology at Soochow University, Suzhou. Her research interest is entity linking.
  • Supported by:
    This work was supported by the National Key Research and Development Program of China under Grant No. 2018AAA0101900, the Natural Science Foundation of Jiangsu Province of China under Grant No. BK2019 1420, the National Natural Science Foundation of China under Grant No. 61632016, the Natural Science Research Project of Jiangsu Higher Education Institution under Grant No. 17KJA5 20003, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the Suda-Toycloud Data Intelligence Joint Laboratory.

PDF (PC)

95

补充材料

1

实体链接是指把文本中的待链接实体链接到一个定义好的知识库中，消除歧义。现有的方法主要式利用“本地”信息（待链接实体的上下文本信息）或“全局”信息（候选实体之间的关系）。但是，这些方法在一定程度上都存在文本不足的问题，尤其是短文本数据。为了获得更加丰富的本地和全局信息，我们提出通过网络搜索引擎（WSE）从网络上获取额外的相关文本，对待链接实体上下文进行补充。基于此，我们提出了两种优化实体链接的方法。第一种方法是将网络搜索结果添加到基于词嵌入的实体链接方法中，扩充待链接实体的上下文本，从而提高链接效果。第二种方法是使用网络文本扩展全局信息，通过额外文本，挑选合适的命名实体加入图模型的构建，从而提高基于图的实体链接方法效果。在处理搜索引擎结果文本时，我们采用了两种方法。一种是应用注意力机制，另一种是使用抽象提取方法。最后，可以结合我们提出的两个优化模型，从本地和全局两方面共同提高实体链接效果。我们在六个现实数据集中的实验结果表明，使用额外的搜索引擎文本扩展本地和全局信息可以有效提高实体链接的F1值。

关键词: 实体链接, 网络搜索引擎, 注意力机制, 摘要提取

Abstract: Entity linking (EL) is the task of determining the identity of textual entity mentions given a predefined knowledge base (KB). Plenty of existing efforts have been made on this task using either "local" information (contextual information of the mention in the text), or "global" information (relations among candidate entities). However, either local or global information might be insufficient especially when the given text is short. To get richer local and global information for entity linking, we propose to enrich the context information for mentions by getting extra contexts from the web through web search engines (WSE). Based on the intuition above, two novel attempts are made. The first one adds web-searched results into an embedding-based method to expand the mention's local information, where we try two different methods to help generate high-quality web contexts:one is to apply the attention mechanism and the other is to use the abstract extraction method. The second one uses the web contexts to extend the global information, i.e., finding and utilizing more extra relevant mentions from the web contexts with a graph-based model. Finally, we combine the two models we propose to use both extended local and global information from the extra web contexts. Our empirical study based on six real-world datasets shows that using extra web contexts to extend the local and the global information could effectively improve the F1 score of entity linking.

Key words: entity linking, web search engine (WSE), attention mechanism, abstract extraction

[1] Yamada I, Shindo H, Takeda H, Takefuji Y. Joint learning of the embedding of words and entities for named entity disambiguation. In Proc. the 20th Conference on Computational Natural Language Learning, August 2016, pp.250-259.
[2] Cai R, Wang H, Zhang J. Learning entity representation for named entity disambiguation. In Proc. the 14th China National Conference on Chinese Computational Linguistics and the 3rd International Symposium on Natural Language Processing Based on Naturally Annotated Big Data, November 2015, pp.267-278.
[3] Blanco R, Ottaviano G, Meij E. Fast and space-efficient entity linking in queries. In Proc. the 8th ACM International Conference on Web Search and Data Mining, February 2015, pp.179-188.
[4] Phan M C, Sun A, Tay Y, Han J, Li C. NeuPL:Attentionbased semantic matching and pairlinking for entity disambiguation. In Proc. the 2017 ACM Conference on Information and Knowledge Management, November 2017, pp.1667-676.
[5] Basile P, Caputo A. Entity linking for tweets. Encyclopedia with Semantic Computing and Robotic Intelligence, 2017, 1(1):Article No. 1630020.
[6] Yih W T, Chang M W, He X, Gao J. Semantic parsing via staged query graph generation:Question answering with knowledge base. In Proc. the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing, July 2015, pp.1321-1331.
[7] Ji H, Grishman R, Dang H T, Griffitt K, Ellis J. Overview of the TAC 2010 knowledge base population track. In Proc. the 3rd Text Analysis Conference, November 2010.
[8] Mcnamee P, Dang H T. Overview of the TAC 2009 knowledge base population track. In Proc. the 2nd Text Analysis Conference, November 2009, pp.111-113.
[9] Hoffmann R, Zhang C, Ling X, Zettlemoyer L S, Weld D S. Knowledge-based weak supervision for information extraction of overlapping relations. In Proc. the 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies, June 2011, pp.541-550.
[10] Wang Y, Li Z, Yang Q, Chen Z, Liu A, Liu G, Zhao L. WebEL:Improving entity linking with extra web contexts. In Proc. the 20th International Conference on Web Information Systems Engineering, November 2019, pp.507-522.
[11] Bunescu R C, Pasca M. Using encyclopedic knowledge for named entity disambiguation. In Proc. the 11th Conference of the European Chapter of the Association for Computational Linguistics, April 2006, pp.9-16.
[12] Mihalcea R, Csomai A. Wikify! Linking documents to encyclopedic knowledge. In Proc. the 16th ACM Conference on Information and Knowledge Management, November 2007, pp.233-242.
[13] Fang W, Zhang J, Wang D, Chen Z, Li M. Entity disambiguation by knowledge and text jointly embedding. In Proc. the 20th SIGNLL Conference on Computational Natural Language Learning, August 2016, pp.260-269.
[14] Nguyen T H, Fauceglia N R, Muro M R, Hassanzadeh O, Gliozzo A, Sadoghi M. Joint learning of local and global features for entity linking via neural networks. In Proc. the 26th International Conference on Computational Linguistics, December 2016, pp.2310-2320.
[15] Luo A, Gao S, Xu Y. Deep semantic match model for entity linking using knowledge graph and text. Procedia Computer Science, 2018, 129:110-114.
[16] Liu C, Li F, Sun X, Han H. Attention-based joint entity linking with entity embedding. Information, 2019, 10(2):Article No. 46.
[17] Cucerzan S. Large-scale named entity disambiguation based on Wikipedia data. In Proc. the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, June 2007, pp.708-716.
[18] Witten I H, Milne D N. An effective, low-cost measure of semantic relatedness obtained from Wikipedia links. In Proc. the 1st AAAI Workshop on Wikipedia and Artificial Intelligence, July 2008, pp.25-30.
[19] Hoffart J, Yosef M A, Bordino I, Füurstenau H, Pinkal M, Spaniol M, Taneva B, Thater S, Weikum G. Robust disambiguation of named entities in text. In Proc. the Conference on Empirical Methods in Natural Language Processing, July 2011, pp.782-792.
[20] Ratinov L A, Roth D, Downey D, Anderson M. Local and global algorithms for disambiguation to Wikipedia. In Proc. the 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies, June 2011, pp.1375-1384.
[21] Alhelbawy A, Gaizauskas R. Graph ranking for collective named entity disambiguation. In Proc. the 52nd Annual Meeting of the Association for Computational Linguistics, June 2014, pp.75-80.
[22] Liu M, Chen L, Liu B, Zheng G, Zhang X. DBpedia-based entity linking via greedy search and adjusted Monte Carlo random walk. ACM Transactions on Information Systems, 2017, 36(2):Article No. 16.
[23] Liu G, Wang Y, Orgun M A, Lim E P. Finding the optimal social trust paths for the selection of trustworthy service providers in complex social networks. IEEE Transactions on Services Computing, 2013, 6(2):152-167.
[24] Liu G, Yan W, Orgun M A. Optimal social trust path selection in complex social networks. In Proc. the 24th AAAI Conference on Artificial Intelligence, July 2010.
[25] Han X, Sun L, Zhao J. Collective entity linking in web text:A graph-based method. In Proc. the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, July 2011, pp.765-774.
[26] Ganea O E, Ganea M, Lucchi A, Eickhoff C, Hofmann T. Probabilistic bag-of-hyperlinks model for entity linking. In Proc. the 25th International Conference on World Wide Web, April 2015, pp.927-938.
[27] Xie T, Wu B, Jia B, Wang B. Graph-ranking collective Chinese entity linking algorithm. Frontiers of Computer Science, 2020, 14(2):291-303.
[28] Liu M, Zhao Y, Qin B, Liu T. Collective entity linking:A random walk-based perspective. Knowledge and Information Systems, 2019, 60(3):1611-1643.
[29] Liu F H, Gu L, Gao Y, Picheny M. Use of statistical Ngram models in natural language generation for machine translation. In Proc. the 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, April 2003, pp.636-639.
[30] Wallach H M. Topic modeling:Beyond bag-of-words. In Proc. the 23rd International Conference on Machine Learning, June 2006, pp.977-984.
[31] Pauls A, Dan K. Faster and smaller N-gram language models. In Proc. the 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies, June 2012, pp.258-267.
[32] Lei K, Deng Y, Zhang B, Shen Y. Open domain question answering with character-level deep learning models. In Proc. the 10th International Symposium on Computational Intelligence and Design, December 2018, pp.30-33.
[33] Gang Z, Zong-Min M A, Kan H M, Niu L Q. Texture feature extraction approach using co-occurrence matrix. Journal of Shenyang University of Technology, 2010, 32(2):192-195. (in Chinese)
[34] Mikolov T, Chen K, Corrado G et al. Efficient estimation of word representations in vector space. In Proc. the 1st International Conference on Learning Representations, May 2013.
[35] Landgraf A J, Bellay J. Word2vec skip-gram with negative sampling is a weighted logistic PCA. arXiv:1705.09755, 2017. https://arxiv.org/abs/1705.09755, March 2020.
[36] Francis-landau M, Durrett G, Klein D. Capturing semantic similarity for entity linking with convolutional neural networks. In Proc. the 2016 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies, June 2016, pp.1256-1261.
[37] Sun Y, Lin L, Tang D, Yang N, Ji Z, Wang X. Modeling mention, context and entity with neural networks for entity disambiguation. In Proc. the 24th International Conference on Artificial Intelligence, July 2015, pp.1333-1339.
[38] Shi W, Zhang S, Zhang Z, Cheng H, Yu J X. Joint embedding in named entity linking on sentence level. arXiv:2002.04936, 2020. ttps://arxiv.org/pdf/2002.04936.pdf, March 2020.
[39] Le P, Titov I. Improving entity linking by modeling latent relations between mentions. In Proc. the 56th Annual Meeting of the Association for Computational Linguistics, July 2018, pp.1595-1604.
[40] Ensan F, Du W. Ad hoc retrieval via entity linking and semantic similarity. Knowledge and Information Systems, 2019, 58(3):551-583.
[41] Lehmann J, Monahan S, Nezda L, Jung A, Shi Y. LCC approaches to knowledge base population at TAC 2010. In Proc. the 3rd Text Analysis Conference, November 2010.
[42] Monahan S, Lehmann J, Nyberg T, Plymale J, Jung A. Cross-lingual cross-document coreference with entity linking. In Proc. the 4th Text Analysis Conference, November 2011.
[43] Cornolti M, Ferragina P, Ciaramita M, Rüd S, Schütze H. SMAPH:A piggyback approach for entity-linking in web queries. ACM Transactions on Information Systems, 2018, 37(1):Article No. 13.
[44] Trani S, Ceccarelli D, Lucchese C, Orlando S, Perego R. SEL:A unified algorithm for entity linking and saliency detection. In Proc. the 2016 ACM Symposium on Document Engineering, September 2016, pp.85-94.
[45] See A, Liu P J, Manning C D. Get to the point:Summarization with pointer-generator networks. In Proc. the 55th Annual Meeting of the Association for Computational Linguistics, July 2017, pp.1073-1083.
[46] Ganea O E, Hofmann T. Deep joint entity disambiguation with local neural attention. In Proc. the 2017 Conference on Empirical Methods in Natural Language Processing, September 2017, pp.2619-2629.
[47] Cheng X, Roth D. Relational inference for Wikification. In Proc. the 2013 Conference on Empirical Methods in Natural Language Processing, October 2013, pp.1787-1796.
[48] Chisholm A, Hachey B. Entity disambiguation with web links. Transactions of the Association for Computational Linguistics, 2015, 3:145-156.
[49] Globerson A, Lazic N, Chakrabarti S, Subramanya A, Ringaard M, Pereira F. Collective entity resolution with multi-focal attention. In Proc. the 54th Annual Meeting of the Association for Computational Linguistics, August 2016, pp.621-631.
[1] Jia-Ke Ge, Yan-Feng Chai, Yun-Peng Chai. WATuning：一种基于注意力机制的深度强化学习的工作负载感知调优系统[J]. 计算机科学技术学报, 2021, 36(4): 741-761.
[2] Chen-Chen Sun, De-Rong Shen. 面向深度实体匹配的混合层次网络[J]. 计算机科学技术学报, 2021, 36(4): 822-838.
[3] Sheng-Luan Hou, Xi-Kun Huang, Chao-Qun Fei, Shu-Han Zhang, Yang-Yang Li, Qi-Lin Sun, Chuan-Qing Wang. 基于深度学习的文本摘要研究综述[J]. 计算机科学技术学报, 2021, 36(3): 633-663.
[4] Yang Liu, Ruili He, Xiaoqian Lv, Wei Wang, Xin Sun, Shengping Zhang. 婴儿的年龄和性别容易被识别吗？[J]. 计算机科学技术学报, 2021, 36(3): 508-519.
[5] Ying Li, Jia-Jie Xu, Peng-Peng Zhao, Jun-Hua Fang, Wei Chen, Lei Zhao. ATLRec:用于跨领域推荐的注意力对抗迁移学习网络[J]. 计算机科学技术学报, 2020, 35(4): 794-808.
[6] Huan-Jing Yue, Sheng Shen, Jing-Yu Yang, Hao-Feng Hu, Yan-Fang Chen. 基于渐进式通道注意力网络的参考图引导超分辨率研究[J]. 计算机科学技术学报, 2020, 35(3): 551-563.
[7] Chun-Yang Ruan, Ye Wang, Jiangang Ma, Yanchun Zhang, Xin-Tian Chen. 基于元路径注意力机制的异构网络对抗式嵌入[J]. 计算机科学技术学报, 2019, 34(6): 1217-1229.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 冯玉琳;. Recursive Implementation of VLSI Circuits[J]. , 1986, 1(2): 72 -82 .
[2] 王选; 吕之敏; 汤玉海; 向阳;. A High Resolution Chinese Character Generator[J]. , 1986, 1(2): 1 -14 .
[3] 陈肇雄; 高庆狮;. A Substitution Based Model for the Implementation of PROLOG——The Design and Implementation of LPROLOG[J]. , 1986, 1(4): 17 -26 .
[4] 黄学东; 蔡莲红; 方棣棠; 迟边进; 周立; 蒋力;. A Computer System for Chinese Character Speech Input[J]. , 1986, 1(4): 75 -83 .
[5] 龚振和;. On Conceptual Model Specification and Verification[J]. , 1987, 2(1): 35 -50 .
[6] S.T.Chanson; 梁路平; A.Kumar;. Throughput Models of CSMA Network with Stations Uniformly Distributed along the Bus[J]. , 1987, 2(4): 243 -264 .
[7] 王建潮; 魏道政;. Reconvergent-Fanout-Oriented Testability Measure[J]. , 1988, 3(1): 16 -28 .
[8] 包丰;. On the Condition for FSM Being a Scrambler[J]. , 1988, 3(1): 70 -74 .
[9] 孙成政; 慈云桂;. A New Method for Describing the AND-OR-Parallel Execution of Logic Programs[J]. , 1988, 3(2): 102 -112 .
[10] 张钹; 张恬; 张建伟; 张铃;. Motion Planning for Robots with Topological Dimension Reduction Method[J]. , 1990, 5(1): 1 -16 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持：support@magtech.com.cn
总访问量：