Elamur Research⌕
☰
home / papers / 2016 / TAXI at SemEval-2016 Task 13: a Taxonomy Induction Method based on Lexico-Syntactic Patterns, Substrings and Focused Crawling

TAXI at SemEval-2016 Task 13: a Taxonomy Induction Method based on Lexico-Syntactic Patterns, Substrings and Focused Crawling

Paper · 2016 · Alexander Panchenko, Stefano Faralli, Eugen Ruppert et al.

Summary

The paper presents TAXI, the system that won every subtask of SemEval-2016 Task 13 (Taxonomy Extraction Evaluation). The task is to build a taxonomy from a list of domain terms. TAXI relies on raw text and avoids structured resources such as WordNet or BabelNet, so it can be applied to new languages and domains. The authors first build large domain-specific corpora by focused crawling. BootCat supplies seed URLs, and a 5-gram Kneser-Ney language model trained on Wikipedia category articles decides which pages are relevant. These crawls supplement Wikipedia, 59G and CommonCrawl. Hypernym candidates come from two sources:

  • a substring score, which captures compound terms;
  • an asymmetric score that merges Hearst-style pattern extractions from PattaMaika, PatternSim and WebISA.

The candidates are pruned to the most salient hypernyms per term. Finally, cycles are broken and disconnected components are attached to the root. The method was tested on English, Dutch, French and Italian across the Food, Environment and Science domains. This fairly simple, data-heavy approach outperformed more complex and knowledge-rich competitors, particularly in the multilingual setting.

Key points

  • Each language–domain pair was crawled for roughly one week on a single 24-core server. This produced 130–800 GB of raw content, which reduced to 2–23 GB of unique plaintext sentences.
  • On the trial data, the substring score kept precision high, reaching 0.91 at 0.29 recall (AUC 0.61). It can only find hypernyms for compound terms.
  • Pattern extraction combined three sources: PattaMaika (UIMA Ruta rules for EN/IT/NL), PatternSim (Unitex transducers for EN/FR) and the WebISA database, from which 108 million English relations were used. Altogether this gave 18 hypernym collections.
  • Human judges rated TAXI first for Dutch, French and Italian, with an average precision of 0.625. For English it placed second with 0.20, behind the substring-based USAAR system at 0.49.
  • Coverage remains limited: 774 of the 1,555 English Food terms were left unattached. The code and resources are released at tudarmstadt-lt.github.io/taxi.

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

 Institutions

🎓Technische Universität DarmstadtDarmstadt · Germany123 works · 36,746 citations🎓University of MannheimMannheim · Germany31 works · 8,538 citations🎓UCLouvainLouvain-la-Neuve · Belgium21 works · 3,831 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 19,503 worksArtificial Intelligence · 18,805 worksArts and Humanities · 17,959 worksLanguage and Linguistics · 17,789 works

Cite

Cite as Alexander Panchenko, Stefano Faralli, Eugen Ruppert, Steffen Remus, Hubert Naets, Cédrick Fairon, Simone Paolo Ponzetto and Chris Biemann. 2016. TAXI at SemEval-2016 Task 13: a Taxonomy Induction Method based on Lexico-Syntactic Patterns, Substrings and Focused Crawling. Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/s16-1206

@inproceedings{panchenko2016taxi,
  title = {{TAXI at SemEval-2016 Task 13: a Taxonomy Induction Method based on Lexico-Syntactic Patterns, Substrings and Focused Crawling}},
  author = {Alexander Panchenko, Stefano Faralli, Eugen Ruppert, Steffen Remus, Hubert Naets, Cédrick Fairon, Simone Paolo Ponzetto and Chris Biemann},
  booktitle = {Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016)},
  year = {2016},
  doi = {10.18653/v1/s16-1206},
  url = {https://doi.org/10.18653/v1/s16-1206}
}

References

34 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Eneko Agirre, Olatz Ansa, Eduard H. Hovy, and David Martı́nez. 2000. Enriching very large ontologies using the WWW. In ECAI Workshop on Ontology Learning, pages 28–33, Berlin, Germany. find
  2. Marco Baroni and Silvia Bernardini. 2004. BootCaT: Bootstrapping corpora and terms from the web. In Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC), pages 1313–1316, Lisbon, Portugal.
  3. Chris Biemann and Martin Riedl. 2013. Text: Now in 2D! a framework for lexical expansion with contextual similarity. Journal of Language Modelling, 1(1):55– 95.
  4. Chris Biemann. 2005. Ontology Learning from Text: A Survey of Methods. LDV Forum, 20(2):75–93. find
  5. Georgeta Bordea, Paul Buitelaar, Stefano Faralli, and Roberto Navigli. 2015. Semeval-2015 task 17: Taxonomy extraction evaluation (TExEval). SemEval-2015, 452(465):902–910. find
  6. Georgeta Bordea, Els Lefever, and Paul Buitelaar. 2016. Semeval-2016 task 13: Taxonomy extraction evaluation (TExEval-2). In Proceedings of the 10th International Workshop on Semantic Evaluation, San Diego, CA, USA. find
  7. Soumen Chakrabarti, Martin van den Berg, and Byron Dom. 1999. Focused crawling: a new approach to topic-specific web resource discovery. Computer Networks, 31(11–16):1623–1640. find
  8. Guillaume Cleuziou and Jose G. Moreno. 2016. QAS-SIT at SemEval-2016 Task 13: On the integration of Semantic Vectors in Pretopological Spaces for Lexical Taxonomy Acquisition. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval 2016). Association for Computational Lingu… find
  9. Stefano Faralli, Giovanni Stilo, and Paola Velardi. 2015. Large scale homophily analysis in twitter using a twixonomy. In Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI 2015, pages 2334–2340, Buenos Aires, Argentina. find
  10. Adriano Ferraresi, Eros Zanchetta, Marco Baroni, and Silvia Bernardini. 2008. Introducing and evaluating ukWaC, a very large web-derived corpus of English. In Proceedings of the 4th Web as Corpus Workshop (WAC-4) Can we beat Google, pages 47–54, Marakech, Morocco. find
  11. Dirk Goldhahn, Thomas Eckart, and Uwe Quasthoff. 2012. Building Large Monolingual Dictionaries at the Leipzig Corpora Collection: From 100 to 200 Languages. In Proceedings of the Eight International Conference on Language Resources and Evaluation (LREC’12), pages 759–765, Istanbul, Turkey.
  12. Gregory Grefenstette. 2015. INRIASAC: Simple hypernym extraction methods. SemEval-2015, pages 911– 914. find
  13. Marti A Hearst. 1992. Automatic acquisition of hyponyms from large text corpora. In Proceedings of the 14th conference on Computational linguistics-Volume 2, pages 539–545.
  14. Carmen Klaussner and Desislava Zhekova. 2011. Lexico-syntactic patterns for automatic ontology building. In Student Research Workshop at RANLP 2011, pages 109–114. find
  15. Peter Kluegl, Martin Toepfer, Philip-Daniel Beck, Georg Fette, and Frank Puppe. 2014. UIMA Ruta: Rapid development of rule-based information extraction applications. Natural Language Engineering, 22(1):1–40. find
  16. Reinhard Kneser and Hermann Ney. 1995. Improved backing-off for m-gram language modeling. In Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on, volume 1, pages 181–184. find
  17. IEEE. Zornitsa Kozareva and Eduard Hovy. 2010. Learning arguments and supertypes of semantic relations using recursive patterns. In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics (ACL), pages 1482–1491, Uppsala, Sweden. find
  18. Els Lefever. 2015. LT3: a multi-modular approach to automatic taxonomy construction. SemEval-2015, pages 944–948. find
  19. George A Miller. 1995. Wordnet: a lexical database for english. Communications of the ACM, 38(11):39–41.
  20. Roberto Navigli and Simone Paolo Ponzetto. 2010. BabelNet: building a very large multilingual semantic network. In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics (ACL), pages 216–225, Uppsala, Sweden.
  21. Michael P. Oakes. 2005. Using Hearst’s rules for the automatic acquisition of hyponyms for mining a pharmaceutical corpus. In RANLP Text Mining Workshop’05, pages 63–67, Borovets, Bulgaria. find
  22. Alexander Panchenko, Olga Morozova, and Hubert Naets. 2012. A semantic similarity measure based on lexico-syntactic patterns. In Jeremy Jancsary, editor, Proceedings of KONVENS 2012, pages 174–178, Vienna, Austria. ÖGAI. Main track: poster presentations. Alexander Panchenko, Hubert Naets, Laetitia… find
  23. Robert Parker, David Graff, Junbo Kong, Ke Chen, and Kazuaki Maeda. 2009. English gigaword fourth edition. find
  24. Linguistic Data Consortium, Philadelphia, USA. Joel Pocostales. 2016. NUIG-UNLP at SemEval-2016 Task 13: A Simple Word Embedding-based Approach for Taxonomy Extraction. In Proceedings of the 10th International Workshop on Semantic Evaluation. Association for Computational Linguistics. find
  25. Steffen Remus and Chris Biemann. 2016. Domain-specific corpus expansion with focused webcrawling. In Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC), Portorož, Slovenia. find
  26. Alan Ritter, Stephen Soderland, and Oren Etzioni. 2009. What is this, anyway: Automatic hypernym discovery. In Proceedings of the 2009 AAAI Spring Symposium on Learning by Reading and Learning to Read, pages 88–93, Palo Alto, California. find
  27. Julian Seitner, Christian Bizer, Kai Eckert, Stefano Faralli, Robert Meusel, Heiko Paulheim, and Simone Ponzetto. 2016. A Large DataBase of Hypernymy Relations Extracted from the Web. In Proceedings of the 10th edition of the Language Resources and Evaluation Conference, Portorož, Slovenia.
  28. Rion Snow, Daniel Jurafsky, and Andrew Y. Ng. 2004. Learning syntactic patterns for automatic hypernym discovery. In Lawrence K. Saul, Yair Weiss, and Léon Bottou, editors, Proc. of NIPS 2004, pages 1297–1304, Cambridge, Mass. find
  29. MIT Press. Rion Snow, Dan Jurafsky, and Andrew Ng. 2006. Semantic taxonomy induction from heterogeneous evidence. In Proceedings of COLING-ACL 2006, pages 801–808, Sydney, Australia. find
  30. Liling Tan, Rohit Gupta, and Josef van Genabith. 2015. USAAR-WLV: hypernym generation with deep neural nets. SemEval-2015, pages 932–937. find
  31. Liling Tan, Francis Bond, and Josef van Genabith. 2016. USAAR at SemEval-2016 Task 13: Hyponym Endocentricity. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval 2016). Association for Computational Linguistics. find
  32. Robert Tarjan. 1972. Depth-first search and linear graph algorithms. SIAM Journal on Computing, 1:146–160. find
  33. Paola Velardi, Stefano Faralli, and Roberto Navigli. 2013. Ontolearn reloaded: A graph-based algorithm for taxonomy induction. Computational Linguistics, 39(3):665–707. Jean-Philippe Vert, Koji Tsuda, and Bernhard Schölkopf. 2004. A primer on kernel methods. In Bernhard Schlkopf, Koji Tsuda, and Je…
  34. Hui Yang and Jamie Callan. 2009. A metric-based framework for automatic taxonomy induction. In Proceedings of the Joint Conference of the 47th Annual Meeting of the Association for Computational Linguistics (ACL) and the 4th International Joint Conference on Natural Language Processing of the Asian… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/panchenko-2016-taxi-semeval-2016-task-13/