Elamur Research⌕
☰
home / papers / 2014 / Entity Linking meets Word Sense Disambiguation: a Unified Approach

Entity Linking meets Word Sense Disambiguation: a Unified Approach

Summary

The paper unifies two related lexical disambiguation tasks: Entity Linking (EL), as well as Word Sense Disambiguation (WSD). They differ primarily in how a mention corresponds to a meaning. In WSD, the lemma exactly matches one of the senses in a dictionary. In EL, the mention may be partial, for example only a footballer's first name, and must point to an entry in an encyclopedic knowledge base. The authors propose Babelfy, a unified graph-based approach built on the multilingual semantic network BabelNet 1.1.1, which combines Wikipedia and WordNet. The method has three steps. First, a semantic signature is computed once for each vertex of the network: random walk with restart is applied, and edge weights are refined using triangles. Next, candidate meanings are selected by partial matching for all text fragments, including overlapping ones. Finally, a densest-subgraph heuristic retains the most coherent joint interpretation of concepts and named entities. On six gold-standard datasets, including multilingual ones, the system achieves state-of-the-art results in both tasks. This indicates that lexicographic and encyclopedic knowledge reinforce each other. Thus, a single model can address both tasks in any language covered by BabelNet.

Key points

  • On KORE50, partial candidate matching includes the correct entity in 98.6% of cases. Exact string matching yields only 42.4%.
  • Ablation on KORE50: without the densest-subgraph heuristic, EL performance drops by 9%, while WSD is almost unaffected. Restricting the graph to entities only (for EL) or concepts only (for WSD) costs 3.4%. Processing individual sentences instead of whole documents also generally degrades performance.
  • The densest subgraph is found using a 2-approximation algorithm. BabelNet on its own, even combined with Personalized PageRank (UKB), does not achieve state-of-the-art results.
  • The networks are sparse: the average number of incident edges per vertex is roughly 10 for WordNet, 50 for BabelNet, and 80 for YAGO2.
  • Evaluation was carried out on, among others, KORE50, SemEval-2007 (task 17 and coarse-grained), and SemEval-2013 task 12 (including Spanish and German). The paper appeared in TACL 2014, and the system is available at babelfy.org.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: AI & machine learning

Topics: Named entity recognition · Word sense disambiguation

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1162/tacl_a_00179

@misc{moro2014entity,
  title = {{Entity Linking meets Word Sense Disambiguation: a Unified Approach}},
  author = {Moro, Andrea and Raganato, Alessandro and Navigli, Roberto},
  year = {2014},
  doi = {10.1162/tacl\_a\_00179},
  url = {https://doi.org/10.1162/tacl_a_00179}
}

References

78 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Eneko Agirre and Aitor Soroa. 2009. Personalizing PageRank for Word Sense Disambiguation. In Proc. of EACL, pages 33–41. find
  2. Eneko Agirre, David Martı́nez, Oier López de Lacalle, and Aitor Soroa. 2006. Two graph-based algorithms for state-of-the-art WSD. In Proc. of EMNLP, pages 585–593. find
  3. Eneko Agirre, Aitor Soroa, and Mark Stevenson. 2010. Graph-based Word Sense Disambiguation of biomedical documents. Bioinformatics, 26(22):2889–2896. find
  4. Eneko Agirre, Oier Lopez de Lacalle, and Aitor Soroa. 2014. Random Walks for Knowledge-Based Word Sense Disambiguation. Computational Linguistics, 40(1):57–84. find our card
  5. Sören Auer, Christian Bizer, Georgi Kobilarov, Jens Lehmann, Richard Cyganiak, and Zachary Ives. 2007. DBpedia: A Nucleus for a Web of Open Data. In Proc. of ISWC/ASWC, pages 722–735. find
  6. Carmen Banea and Rada Mihalcea. 2011. Word Sense Disambiguation with multilingual features. In Proc. of IWCS, pages 25–34. find
  7. Christoph Böhm, Gerard de Melo, Felix Naumann, and Gerhard Weikum. 2012. LINDA: distributed web-of-data-scale entity matching. In Proc. of CIKM, pages 2104–2108. find
  8. Samuel Brody and Mirella Lapata. 2009. Bayesian Word Sense Induction. In Proc. of EACL, pages 103–111. find our card
  9. Razvan C. Bunescu and Marius Pasca. 2006. Using encyclopedic knowledge for named entity disambiguation. In Proc. of EACL, pages 9–16. find our card
  10. Yee Seng Chan, Hwee Tou Ng, and Zhi Zhong. 2007. NUS-PT: Exploiting Parallel Texts for Word Sense Disambiguation in the English All-Words Tasks. In Proc. of SemEval-2007, pages 253–256. find
  11. Moses Charikar. 2000. Greedy approximation algorithms for finding dense components in a graph. In Proc. of APPROX, pages 84–95. find
  12. Xiao Cheng and Dan Roth. 2013. Relational Inference for Wikification. In Proc. of EMNLP, pages 1787– 1796. find
  13. Marco Cornolti, Paolo Ferragina, and Massimiliano Ciaramita. 2013. A framework for benchmarking entity-annotation systems. In Proc. of WWW, pages 249– 260. find
  14. Silviu Cucerzan. 2007. Large-Scale Named Entity Disambiguation Based on Wikipedia Data. In Proc. of EMNLP-CoNLL, pages 708–716. find
  15. Joachim Daiber, Max Jakob, Chris Hokamp, and Pablo N. Mendes. 2013. Improving efficiency and accuracy in multilingual entity extraction. In Proc. of I-Semantics, pages 121–124. find
  16. Leon Derczynski, Diana Maynard, Niraj Aswani, and Kalina Bontcheva. 2013. Microblog-genre noise and impact on semantic annotation accuracy. In Proc. of Hypertext, pages 21–30. find
  17. Antonio Di Marco and Roberto Navigli. 2013. Clustering and Diversifying Web Search Results with Graph-Based Word Sense Induction. Computational Linguistics, 39(3):709–754. find our card
  18. Nicolai Erbs, Torsten Zesch, and Iryna Gurevych. 2011. Link discovery: A comprehensive analysis. In Proc. of ICSC, pages 83–86. find
  19. Oren Etzioni, Michele Banko, and Michael J Cafarella. 2006. Machine Reading. In Proc. of AAAI, pages 1517–1519. find
  20. Uriel Feige, Guy Kortsarz, and David Peleg. 1999. The dense k-subgraph problem. Algorithmica, 29:2001. find
  21. Christiane Fellbaum. 1998. WordNet: An Electronic Lexical Database. MIT Press. find
  22. Paolo Ferragina and Ugo Scaiella. 2010. TAGME: On-the-fly Annotation of Short Text Fragments (by Wikipedia Entities). In Proc. of CIKM, pages 1625– 1628. find our card
  23. Paolo Ferragina and Ugo Scaiella. 2012. Fast and Accurate Annotation of Short Texts with Wikipedia Pages. IEEE Software, 29(1):70–75. find
  24. Weiwei Guo and Mona T. Diab. 2010. Combining Orthogonal Monolingual and Multilingual Sources of Evidence for All Words WSD. In Proc. of ACL, pages 1542–1551. find
  25. Yoan Gutiérrez, Yenier Castañeda, Andy González, Rainel Estrada, Dennys D. Piug, Jose I. Abreu, Roger Pérez, Antonio Fernández Orquı́n, Andrés Montoyo, Rafael Muñoz, and Franc Camara. 2013. UMCC DLSI: Reinforcing a Ranking Algorithm with Sense Frequencies and Multidimensional Semantic Resourc… find
  26. Ben Hachey, Will Radford, Joel Nothman, Matthew Honnibal, and James R. Curran. 2013. Evaluating Entity Linking with Wikipedia. Artificial Intelligence, 194:130–150. find
  27. Sherzod Hakimov, Salih Atilay Oto, and Erdogan Dogdu. 2012. Named entity recognition and disambiguation using linked data and graph-based centrality scoring. In Proc. of SWIM, pages 4:1–4:7. find
  28. Taher H. Haveliwala. 2002. Topic-sensitive PageRank. In Proc. of WWW, pages 517–526. find
  29. Johannes Hoffart, Mohamed Amir Yosef, Ilaria Bordino, Hagen Fürstenau, Manfred Pinkal, Marc Spaniol, Bilyana Taneva, Stefan Thater, and Gerhard Weikum. 2011. Robust disambiguation of named entities in text. In Proc. of EMNLP, pages 782–792. find
  30. Johannes Hoffart, Stephan Seufert, Dat Ba Nguyen, Martin Theobald, and Gerhard Weikum. 2012. KORE: keyphrase overlap relatedness for entity disambiguation. In Proc. of CIKM, pages 545–554. find
  31. Johannes Hoffart, Fabian M Suchanek, Klaus Berberich, and Gerhard Weikum. 2013. YAGO2: A spatially and temporally enhanced knowledge base from Wikipedia. Artificial Intelligence, 194:28–61. find
  32. Eduard H. Hovy, Roberto Navigli, and Simone P. Ponzetto. 2013. Collaboratively built semi-structured content and Artificial Intelligence: The story so far. Artificial Intelligence, 194:2–27. find
  33. Samir Khuller and Barna Saha. 2009. On finding dense subgraphs. In Proc. of ICALP, pages 597–608. find
  34. Sayali Kulkarni, Amit Singh, Ganesh Ramakrishnan, and Soumen Chakrabarti. 2009. Collective Annotation of Wikipedia Entities in Web Text. In Proc. of KDD, pages 457–466. find
  35. Els Lefever and Véronique Hoste. 2010. Semeval-2010 task 3: Cross-lingual Word Sense Disambiguation. In Proc. of SemEval-2010, pages 15–20. find
  36. Els Lefever and Véronique Hoste. 2013. SemEval-2013 Task 10: Cross-lingual Word Sense Disambiguation. In Proc. of SemEval-2013, pages 158–166. find
  37. Els Lefever, Véronique Hoste, and Martine De Cock. 2011. Parasense or how to use parallel corpora for Word Sense Disambiguation. In Proc. of ACL-HLT, pages 317–322. find
  38. Michael E. Lesk. 1986. Automatic Sense Disambiguation Using Machine Readable Dictionaries: How to Tell a Pine Cone from an Ice Cream Cone. In Proc. of the International Conference on Systems Documentation, pages 24–26. find
  39. Nadine Ludwig and Harald Sack. 2011. Named entity recognition for user-generated tags. In Proc. of DEXA, pages 177–181. find
  40. Suresh Manandhar, Ioannis P. Klapaftis, Dmitriy Dligach, and Sameer S. Pradhan. 2010. SemEval-2010 task 14: Word sense induction & disambiguation. In Proc. of SemEval-2010, pages 63–68. find our card
  41. Steve L. Manion and Raazesh Sainudiin. 2013. DAE-BAK!: Peripheral Diversity for Multilingual Word Sense Disambiguation. In Proc. of SemEval-2013, pages 250–254. find
  42. Pablo N. Mendes, Max Jakob, Andrés Garcı́a-Silva, and Christian Bizer. 2011. DBpedia spotlight: shedding light on the web of documents. In Proc. of I-Semantics, pages 1–8. find
  43. Rada Mihalcea and Andras Csomai. 2007. Wikify!: linking documents to encyclopedic knowledge. In Proc. of CIKM, pages 233–242. find
  44. Rada Mihalcea and Dan I Moldovan. 2001. Extended WordNet: Progress report. In Proc. of NAACL Workshop on WordNet and Other Lexical Resources, pages 95–100. find
  45. Rada Mihalcea. 2005. Unsupervised large-vocabulary word sense disambiguation with graph-based algorithms for sequence data labeling. In Proc. of HLT/EMNLP, pages 411–418. find our card
  46. George A. Miller, Claudia Leacock, Randee Tengi, and Ross T. Bunker. 1993. A semantic concordance. In Proc. of HLT, pages 303–308. find our card
  47. Tristan Miller, Chris Biemann, Torsten Zesch, and Iryna Gurevych. 2012. Using Distributional Similarity for Lexical Expansion in Knowledge-based Word Sense Disambiguation. In Proc. of COLING, pages 1781– 1796. find
  48. David Milne and Ian H. Witten. 2008. Learning to link with Wikipedia. In Proc. of CIKM, pages 509–518. find
  49. Andrea Moro and Roberto Navigli. 2013. Integrating Syntactic and Semantic Analysis into the Open Information Extraction Paradigm. In Proc. of IJCAI, pages 2148–2154. find
  50. Roberto Navigli and Mirella Lapata. 2010. An Experimental Study of Graph Connectivity for Unsupervised Word Sense Disambiguation. TPAMI, 32(4):678–692. find
  51. Roberto Navigli and Simone Paolo Ponzetto. 2012a. BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artificial Intelligence, 193:217–250. find
  52. Roberto Navigli and Simone Paolo Ponzetto. 2012b. Joining forces pays off: Multilingual Joint Word Sense Disambiguation. In Proc. of EMNLP, pages 1399– 1410. find
  53. Roberto Navigli, Kenneth C. Litkowski, and Orin Hargraves. 2007. SemEval-2007 Task 07: Coarse-Grained English All-Words Task. In Proc. of SemEval-2007, pages 30–35. find our card
  54. Roberto Navigli, David Jurgens, and Daniele Vannella. 2013. SemEval-2013 Task 12: Multilingual Word Sense Disambiguation. In Proc. of SemEval-2013, pages 222–231. find
  55. Roberto Navigli. 2008. A structural approach to the automatic adjudication of word sense disagreements. Natural Language Engineering, 14(4):293–310. find
  56. Roberto Navigli. 2009. Word Sense Disambiguation: A survey. ACM Computing Surveys, 41(2):1–69. find
  57. Roberto Navigli. 2012. A Quick Tour of Word Sense Disambiguation, Induction and Related Approaches. In Proc. of SOFSEM, pages 115–129. find
  58. Hwee Tou Ng and Hian Beng Lee. 1996. Integrating multiple knowledge sources to disambiguate word sense: An exemplar-based approach. In Proc. of ACL, pages 40–47. find our card
  59. Mohammad Taher Pilehvar and Roberto Navigli. 2014. A Large-scale Pseudoword-based Evaluation Framework for State-of-the-Art Word Sense Disambiguation. Computational Linguistics. find
  60. Simone P. Ponzetto and Roberto Navigli. Knowledge-rich Word Sense Disambiguation rivaling supervised system. In Proc. of ACL, pages 1522–1531. find
  61. Sameer S. Pradhan, Edward Loper, Dmitriy Dligach, and Martha Palmer. 2007. SemEval-2007 task 17: English lexical sample, SRL and all words. In Proc. of SemEval-2007, pages 87–92. Association for Computational Linguistics. find our card
  62. Delip Rao, Paul McNamee, and Mark Dredze. 2013. Entity Linking: Finding Extracted Entities in a Knowledge Base. In Multi-source, Multilingual Information Extraction and Summarization, Theory and Applications of Natural Language Processing, pages 93–115. Springer Berlin Heidelberg. find
  63. Lev-Arie Ratinov, Dan Roth, Doug Downey, and Mike Anderson. 2011. Local and Global Algorithms for Disambiguation to Wikipedia. In Proc. of ACL, pages 1375–1384. find
  64. Lenhart K. Schubert. 2006. Turing’s dream and the knowledge challenge. In Proc. of NCAI, pages 1534– 1538. find
  65. Didier Schwab, Andon Tchechmedjiev, Jérôme Goulian, Mohammad Nasiruddin, Gilles Sérasset, and Hervé Blanchon. 2013. GETALP System: Propagation of a Lesk Measure through an Ant Colony Algorithm. In Proc. of SemEval-2013, pages 232–240. find
  66. Hui Shen, Razvan Bunescu, and Rada Mihalcea. 2013. Coarse to Fine Grained Sense Disambiguation in Wikipedia. In Proc. of *SEM, pages 22–31. find
  67. Ravi Sinha and Rada Mihalcea. 2007. Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity. In Proc. of ICSC, pages 363–369. find
  68. Benjamin Snyder and Martha Palmer. 2004. The English all-words task. In Proc. of Senseval-3, pages 41–43. find
  69. Carlo Strapparava and Alessandro Valitutti. 2004. Word-Net Affect: an Affective Extension of WordNet. In Proc. of LREC, pages 1083–1086. find
  70. Hanghang Tong, Christos Faloutsos, and Jia-Yu Pan. 2006. Fast Random Walk with Restart and Its Applications. In Proc. of ICDM, pages 613–622. find
  71. Kristina Toutanova, Dan Klein, Christopher D. Manning, and Yoram Singer. 2003. Feature-rich part-of-speech tagging with a cyclic dependency network. In Proc. of NAACL-HLT, pages 173–180. find our card
  72. George Tsatsaronis, Michalis Vazirgiannis, and Ion Androutsopoulos. 2007. Word Sense Disambiguation with Spreading Activation Networks Generated from Thesauri. In Proc. of IJCAI, pages 1725–1730. find
  73. Tim Van de Cruys and Marianna Apidianaki. 2011. Latent Semantic Word Sense Induction and Disambiguation. In Proc. of ACL, pages 1476–1485. find
  74. Daniele Vannella, David Jurgens, Daniele Scarfini, 2010. find
  75. Domenico Toscani, and Roberto Navigli. 2014. Validating and Extending Semantic Knowledge Bases using Video Games with a Purpose. In Proc. of ACL. find
  76. Duncan J. Watts and Steven H. Strogatz. 1998. Collective dynamics of ’small-world’ networks. Nature, 393(6684):409–10. find
  77. Zhi Zhong and Hwee Tou Ng. 2010. It Makes Sense: A Wide-Coverage Word Sense Disambiguation System for Free Text. In Proc. of ACL (Demo), pages 78–83. find
  78. Amal Zouaq, Michel Gagnon, and Benoit Ozell. 2009. A SUMO-based Semantic Analysis for Knowledge Extraction. In Proc of LTC. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/moro-2014-entity-linking-meets-word-sense/