Entity Disambiguation Based on Parse Tree Neighbours on Graph Attention Network

  • Kexuan Xin
  • Wen HuaEmail author
  • Yu Liu
  • Xiaofang Zhou
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11881)


Entity disambiguation (ED) aims to link textual mentions in a document to the correct named entities in a knowledge base (KB). Although global ED model usually outperforms local model by collectively linking mentions based on the topical coherence assumption, it may still incur incorrect entity assignment when a document contains multiple topics. Therefore, we propose to extract global features locally, i.e., among a limited number of neighbouring mentions, to combine the respective superiority of both models. In particular, we derive mention neighbours according to the syntactic distance on a dependency parse tree, and propose a tree connection method CoSimTC to measure the cross-tree distance between mentions. Besides, we extend the Graph Attention Network (GAT) to integrate both local and global features to produce a discriminative representation for each candidate entity. Our experimental results on five widely-adopted public datasets demonstrate better performance compared with state-of-the-art approaches.


Entity linking Dependency parse tree Cross-sentence distance Graph Attention Network 


  1. 1.
    Alhelbawy, A., Gaizauskas, R.: Graph ranking for collective named entity disambiguation. ACL 2, 75–80 (2014)Google Scholar
  2. 2.
    Cao, Y., Hou, L., Li, J., Liu, Z.: Neural collective entity linking. In: COLING, pp. 675–686 (2018)Google Scholar
  3. 3.
    Cheng, X., Roth, D.: Relational inference for wikification. In: EMNLP, pp. 1787–1796 (2013)Google Scholar
  4. 4.
    Cucerzan, S.: Large-scale named entity disambiguation based on wikipedia data. In: EMNLP-CoNLL (2007)Google Scholar
  5. 5.
    Dredze, M., McNamee, P., Rao, D., Gerber, A., Finin, T.: Entity disambiguation for knowledge base population. In: COLING, pp. 277–285 (2010)Google Scholar
  6. 6.
    Francis-Landau, M., Durrett, G., Klein, D.: Capturing semantic similarity for entity linking with convolutional neural networks. In: NAACL, pp. 1256–1261 (2016)Google Scholar
  7. 7.
    Gabrilovich, E., Ringgaard, M., Subramanya, A.: FACC1: freebase annotation of clueweb corpora., Cited by 5 (2013)
  8. 8.
    Ganea, O.E., Ganea, M., Lucchi, A., Eickhoff, C., Hofmann, T.: Probabilistic bag-of-hyperlinks model for entity linking. In: WWW, pp. 927–938 (2016)Google Scholar
  9. 9.
    Ganea, O.E., Hofmann, T.: Deep joint entity disambiguation with local neural attention. In: EMNLP, pp. 2619–2629 (2017)Google Scholar
  10. 10.
    Guo, Z., Barbosa, D.: Robust entity linking via random walks. In: CIKM, pp. 499–508. ACM (2014)Google Scholar
  11. 11.
    Guo, Z., Barbosa, D.: Robust named entity disambiguation with random walks. Semant. Web 9(4), 459–479 (2018)CrossRefGoogle Scholar
  12. 12.
    Gupta, N., Singh, S., Roth, D.: Entity linking via joint encoding of types, descriptions, and context. In: EMNLP, pp. 2681–2690 (2017)Google Scholar
  13. 13.
    Hachey, B., Radford, W., Curran, J.R.: Graph-based named entity linking with wikipedia. In: Bouguettaya, A., Hauswirth, M., Liu, L. (eds.) WISE 2011. LNCS, vol. 6997, pp. 213–226. Springer, Heidelberg (2011). Scholar
  14. 14.
    Han, X., Sun, L., Zhao, J.: Collective entity linking in web text: a graph-based method. In: SIGIR, pp. 765–774. ACM (2011)Google Scholar
  15. 15.
    He, Z., Liu, S., Li, M., Zhou, M., Zhang, L., Wang, H.: Learning entity representation for entity disambiguation. In: ACL(Short Papers), vol. 2, pp. 30–34 (2013)Google Scholar
  16. 16.
    Hoffart, J., et al.: Robust disambiguation of named entities in text. In: EMNLP, pp. 782–792 (2011)Google Scholar
  17. 17.
    Hua, W., Wang, Z., Wang, H., Zheng, K., Zhou, X.: Short text understanding through lexical-semantic analysis. In: 2015 IEEE 31st International Conference on Data Engineering, pp. 495–506, April 2015Google Scholar
  18. 18.
    Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. In: ICLR (2016)Google Scholar
  19. 19.
    Le, P., Titov, I.: Improving entity linking by modeling latent relations between mentions. In: ACL, pp. 1595–1604 (2018)Google Scholar
  20. 20.
    Lehmann, J., Monahan, S., Nezda, L., Jung, A., Shi, Y.: LCC approaches to knowledge base population at TAC 2010. In: TAC (2010)Google Scholar
  21. 21.
    Liu, X., Li, Y., Wu, H., Zhou, M., Wei, F., Lu, Y.: Entity linking for tweets. In: ACL, vol. 1, pp. 1304–1311 (2013)Google Scholar
  22. 22.
    Mihalcea, R., Csomai, A.: Wikify!: linking documents to encyclopedic knowledge. In: CIKM, pp. 233–242. ACM (2007)Google Scholar
  23. 23.
    Milne, D., Witten, I.H.: Learning to link with wikipedia. In: CIKM, pp. 509–518. ACM (2008)Google Scholar
  24. 24.
    Mueller, D., Durrett, G.: Effective use of context in noisy entity linking. In: EMNLP, pp. 1024–1029 (2018)Google Scholar
  25. 25.
    Nie, F., Cao, Y., Wang, J., Lin, C.Y., Pan, R.: Mention and entity description co-attention for entity disambiguation. In: AAAI (2018)Google Scholar
  26. 26.
    Pappu, A., Blanco, R., Mehdad, Y., Stent, A., Thadani, K.: Lightweight multilingual entity extraction and linking. In: WSDM, pp. 365–374. ACM (2017)Google Scholar
  27. 27.
    Pennington, J., Socher, R., Manning, C.: Glove: Global vectors for word representation. In: EMNLP, pp. 1532–1543 (2014)Google Scholar
  28. 28.
    Phan, M.C., Sun, A., Tay, Y., Han, J., Li, C.: NeuPL: Attention-based semantic matching and pair-linking for entity disambiguation. In: CIKM, pp. 1667–1676. ACM (2017)Google Scholar
  29. 29.
    Ratinov, L., Roth, D., Downey, D., Anderson, M.: Local and global algorithms for disambiguation to wikipedia. In: ACL, pp. 1375–1384 (2011)Google Scholar
  30. 30.
    Vaswani, A., et al.: Attention is all you need. In: NIPS, pp. 6000–6010 (2017)Google Scholar
  31. 31.
    Veličković, P., et al.: Graph attention networks. In: ICLR (2018)Google Scholar
  32. 32.
    Yamada, I., Shindo, H., Takeda, H., Takefuji, Y.: Joint learning of the embedding of words and entities for named entity disambiguation. In: CoNLL, p. 250 (2016)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Information Technology and Electrical EngineeringThe University of QueenslandBrisbaneAustralia

Personalised recommendations