Elamur Research⌕
☰
home / papers / 2006 / A Preliminary Study for Building the Basque PropBank

A Preliminary Study for Building the Basque PropBank

Summary

The authors propose a way to add one more layer to the Basque corpus EPEC, which already has syntactic annotation: semantic roles. The methodology combines three resources: the PropBank annotation scheme together with the VerbNet lexicon, their own database of syntactic-semantic subcategorization frames of Basque verbs (Aldezabal, 2004), and the Basque dependency treebank. To test the approach, they took three very frequent verbs: esan, eskatu and adierazi. For each, they compiled lexical entries with rolesets and framesets, and then manually annotated all their arguments and adjuncts in the corpus. Unlike the English PropBank, the frames explicitly specify the case suffixes and postpositions that realize each argument. The study showed that the PropBank model suits agglutinative Basque well. Discrepancies over what counts as an argument and what as an adjunct turned out to be minor and easy to adjust. In addition, the authors found that simple case-based heuristics make it possible to automatically pre-annotate a substantial share of the examples. This lowers the cost of building a semantic role resource for a low-resource language and makes it compatible with other PropBank projects for cross-lingual research.

Key points

  • EPEC is a reference corpus of Basque of about 300,000 words. Roughly a third of the texts come from the Statistical Corpus of 20th Century Basque, the rest from the newspaper Euskaldunon Egunkaria. The corpus is already annotated morphosyntactically, syntactically (dependencies) and with WordNet word senses.
  • The corpus contains 622 verbs. The 10 most frequent account for about 29.95% of all verb occurrences. The light and modal verbs egin and izan were excluded from the pilot study.
  • The stages of the methodology: verb selection, a preliminary lexical entry, comparison with English equivalents in PropBank/VerbNet, corpus annotation, and analysis of the results with a return to the second stage.
  • A general heuristic: the ergative in all occurrences of the studied verbs receives Arg0. For esan, which has two rolesets, the completive and the instrumental unambiguously determine the sense and the role. This fully annotates about 82% of occurrences, leaving only 18% for manual disambiguation.
  • For eskatu and adierazi, which have one roleset each, the accuracy of the heuristics is almost 100%. However, the authors stress that there has been no evaluation on new data yet and that all examples will be checked by a human anyway.

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: Linguistics & languages

Topics: Semantic role labeling

Cited by

5 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 4 not yet.

Cite

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

@inproceedings{agirre2006preliminary,
  title = {{A Preliminary Study for Building the Basque PropBank}},
  author = {Agirre, Eneko and Aldezabal, Izaskun and Etxeberria, Jone and Pociello, Eli},
  booktitle = {Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)},
  year = {2006},
  doi = {10.63317/53p3znx7mavz},
  url = {https://doi.org/10.63317/53p3znx7mavz}
}

References

21 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Aduriz, I., Alegria, I., Arriola, J., Artola, X., Zubillaga, X., Díaz de Ilarraza, A., Ezeiza, N. (1994). EUSLEM: un lematizador/etiquetador de textos en euskera. Actas del X congreso de la Sociedad Española para el Procesamiento del Lenguaje Natural (SEPLN). Grupo: 6 Análisis de corpus. Córdoba. find
  2. Aduriz, I., Agirre, E., Aldezabal, I., Alegria, I., Ansa, O., Arregi, X., Arriola, J., Artola, X., Zubillaga, X., Díaz de Ilarraza, A., Ezeiza, N., Gojenola, K., Maritxalar, M., Oronoz, M., Sarasola, K., Soroa, A., Urizar, R. (1998) A framework for the automatic processing of Basque. In Proceedings… find
  3. Aduriz, I., Aranzabe, M., Arriola, J., Atutxa, A., Díaz de Ilarraza, A., Garmendia, A., Oronoz, M. (2003). Construction of a Basque Dependency Treebank. In TLT 2003. Second Workshop on Treebanks and Linguistic Theories. Vaxjo, Sweden, November 14-15. find
  4. Agirre, E., Alegria, I., Arregi, X., Artola, X., Zubillaga, X., Díaz de Ilarraza, A., Maritxalar, M., Sarasola, K. (1992). Xuxen: A Spelling Checker/Corrector for Basque based in Two-Level Morphology. In Proceedings of ANLP'92, 119-125. Povo Trento. find
  5. Agirre E., Aldezabal I., Pociello E. (2003). A pilot study of English Selectional Preferences and their Cross-Lingual Compatibility with Basque. International Conference on Text Speech and Dialogue. Czech Republic, pp. 12-19. find
  6. Agirre, E., Aldezabal, I., Etxeberria, J., Izagirre, I., Mendizábal, K., Pociello, E., Quintian, M. (2006a). A methodology for the joint development of the Basque WordNet and Semcor. In Proceedings of the 5th International Conference on Language Resources and Evaluations (LREC). Genoa, Italy. find
  7. Agirre, E., Aldezabal, I., Pociello, E. (2006b), Lexicalization and multiword expressions in the Basque WordNet. In Proceedings of Third International WordNet Conference. Jeju Island, Korea. find
  8. Aldezabal, I., Aranzabe, M., Atutxa, A., Gojenola, K., Oronoz, M., Sarasola, K. (2003). Application of finite-state transducers to the acquisition of verb subcategorization information. Natural Language Engineering, Volume 9, pp 39-48, ISSN: 1351-3249. Cambridge University Press. find
  9. Aldezabal, I. (2004). Aditz-azpikategorizazioaren azterketa. 100 aditzen azterketa zehatza, Levin (1993) oinarri harturik eta metodo automatikoak baliatuz. Doktorego txostena. Euskal Filologia saila. Leioa. find
  10. Aranzabe, M., Arriola, J., Atutxa, A., Balza, I., Uria, L. (2003). Guía para la anotación sintáctica manual de Eus3LB (corpus del euskera anotado a nivel sintáctico, semántico y pragmático). UPV/EHU/LSI/TR 13-2003. find
  11. Baker, C.F., Fillmore, C.J., Lowe, J.B. (1998). The Berkeley FrameNet project. In Proceedings of the COLING-ACL. Montreal, Canada. find our card
  12. Civit M., Aldezabal I., Pociello E., Taulé M., Aparicio J. and Màrquez L. (2005a). 3LB-LEX: léxico verbal con frames sintáctico-semánticos. In XXI Congreso de la SEPLN. Granada, Spain. find
  13. Civit, M., Castelvi j., Morante, R., Oliver, A., Aparicio, J. (2005b). 4LEX: a Multilingual Lexical Resource. In Croos-Language Knowledge Induction Workshop. EuroLAN Summer School. Babes,-Bolyai University, Cluj-Napoca, Romania. find
  14. Hajiˇc, J., Panevová, J., Urešová, Z., Bémová, A., Koláˇrová, V., Pajas, P. (2003). PDT-VALLEX: Creating a Largecoverage Valency Lexicon for Treebank Annotation. In Proceedings of The Second Workshop on Treebanks and Linguistic Theories. Sweden, pp. 57–68. find
  15. Kingsbury, P., Palmer, M. (2002). From Treebank to PropBank. In Proceedings of the 3rd International Conference on Language Resources and Evaluation (LREC-2002). Las Palmas, Spain. find our card
  16. Kipper, K., Palmer, M., Rambow, O. (2002). Extending PropBank with VerbNet Semantic Predicates. In Workshop on Applied Interlinguas, held in conjunction with AMTA-2002. Tiburon, CA. find
  17. Lersundi, M. (2005). Ezagutza-base lexikala eraikitzeko Euskal Hiztegiko definizioen azterketa sintaktiko-semantikoa. Hitzen arteko erlazio lexiko-semantikoak: definizio-patroiak, eratorpena eta postposizioak. Euskal Filologia Saila. Leioa. find
  18. Levin, B. (1993). English Verb Classes and Alternations. A preliminary Investigation. Chicago and London. The University of Chicago Press. find
  19. Marcus, M. (1994). The Penn TreeBank: A revised corpus design for extracting predicate argument structure. In Proceedings of the ARPA Human Language Technology Workshop. Princeton, NJ. find
  20. Palmer, M., Xue, N. (2003). Annotating the Propositions in the Penn Chinese Treebank. In Proceedings of the Second Sighan Workshop, Sapporo, Japan. find
  21. Palmer, M., Gildea, D., Kingsbury, P. (2005). The Proposition Bank: A Corpus Annotated with Semantic Roles. In Computational Linguistics Journal. 31:1. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/agirre-2006-preliminary-study-building-basque-propbank/