Are the existing training corpora unnecessarily large?

• Autores: Miguel Ballesteros, Jesús Herrera, Virginia Francisco, Pablo Gervás Gómez-Navarro
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 48, 2012, págs. 21-27
• Idioma: inglés
• Títulos paralelos:
  • ¿Son los corpora de dependencias innecesariamente grandes?
• Enlaces
  • Texto completo
• Resumen
  • español

    El tamaño de los corpora de entrenamiento ha sido siempre uno de los cuellos de botella de los analizadores de dependencias, tanto en términos de optimización como en términos de precisión. En previos estudios nos dimos cuenta que los corpora pueden contener proporciones significativas de datos redundantes al nivel de árboles sintácticos. Dado que el desarrollo de estos tipos de corpora requiere un gran esfuerzo consideramos que un proceso apropiado para seleccionar las frases que se incluyen en el producto final pueden proporcionar sistemas entrenados con los mismos resultados (o incluso mejores) utilizando menor cantidad de frases. Este argumento se demuestra en el estudio llevado a cabo que se expone en este artículo.

  • English

    This paper addresses the problem of optimizing the training treebank data because the size and quality of the data has always been a bottleneck for the purposes of training. In previous studies we realized that current corpora used for training machine learning–based dependency parsers contain a significant proportion of redundant information at the syntactic structure level. Since the development of such training corpora involves a big effort, we argue that an appropriate process for selecting the sentences to be included in them can result in having parsing models as accurate as the ones given when training with bigger – non optimized corpora (or alternatively, bigger accuracy for an equivalent annotation effort). This argument is supported by the results of the study we carried out, which is presented in this paper. Therefore, this paper demonstrates that the training corpora contain more information than needed for training accurate data–driven dependency parsers.

    Patrocinador/es: This research is funded by the Spanish Ministry of Education and Science (TIN2009-14659-C03-01 Project), Univers

• Referencias bibliográficas
  • Abeillé, Anne, editor. 2003. Treebanks: Building and Using Parsed Corpora, volume 20 of Text, Speech and Language Technology. Kluwer Academic...
  • Afonso, Susana, Eckhard Bick, Renato Haber, and Diana Santos. 2002. Floresta sintá(c)tica: A treebank for Portuguese. In LREC 2002.
  • Ballesteros, Miguel, Jesús Herrera, Virginia Francisco, and Pablo Gervás. 2010. Improving Parsing Accuracy for Spanish using Maltparser. SEPLN,...
  • Böhmová, A., J. Hajic, E. Hajicová, and B. Hladká. 2003. The PDT: a 3-level annotation scenario. In Abeillé (Abeillé, 2003), chapter 7.
  • Bosco, Cristina, Simonetta Montemagni, Alessandro Mazzei, Vincenzo Lombardo, Felice dell'Orletta, Alessandro Lenci, Leonardo Lesmo, Giuseppe...
  • Brants, Sabine, Stefanie Dipper, Silvia Hansen, Wolfgang Lezius, and George Smith. 2002. The tiger treebank. In Proceedings of the Workshop...
  • Buchholz, Sabine and Erwin Marsi. 2006. Conll-x shared task on multilingual dependency parsing. In CoNLL-X '06: Proceedings of the Tenth...
  • Chen, Keh-Jiann, Chi-Ching Luo, Ming-Chung Chang, Feng-Yi Chen, Chao-Jan Chen, Chu-Ren Huang, and Zhao-Ming Gao. 2003. Sinica treebank: Design...
  • Dzeroski, S., T. Erjavec, N. Ledinek, P. Pajas, Z. Zabokrtsky, and A. Zele. 2006. Towards a Slovene dependency treebank. In In Proc. Int....
  • Hajic, Jan, Otakar Smrz, Petr Zemánek, Jan Snaidauf, and Emanuel Beska. 2004. Prague Arabic dependency treebank: Development in data and tools....
  • Herrera, J. and P. Gervás. 2008. Towards a Dependency Parser for Greek Using a Small Training Data Set. Journal of the Spanish Society for...
  • Kawata, Y. and J. Bartels. 2000. Stylebook for the Japanese treebank in VERBMOBIL. Verbmobil-Report 240, Seminar für Sprachwissenschaft, Universität...
  • Kromann, Matthias T. 2003. The Danish dependency treebank and the underlying linguistic theory. Växjö, Sweden.
  • Nilsson, Jens, Johan Hall, and Joakim Nivre. 2005. MAMBA meets TIGER: Reconstructing a Swedish treebank from antiquity. In Proc. of the NODALIDA...
  • Nivre, Joakim, Johan Hall, Jens Nilsson, Atanas Chanev, Gülsen Eryigit, Sandra Kbler, Svetoslav Marinov, and Erwin Marsi. 2007. Maltparser:...
  • Oflazer, Kemal, Bilge Say, Dilek Zeynep Hakkani-Tür, and Gökhan Tür. 2003. Building a Turkish treebank. In Abeillé (Abeillé, 2003), chapter...
  • Palomar, M., M. Civit, A. Díaz, L. Moreno, E. Bisbal, M. Aranzabe, A. Ageno, M.A. Martí, and Navarro. 2004. 3lb: Construcción de una base...
  • Prokopidis, P., E. Desypri, M. Koutsombogera, H. Papageorgiou, and S. Piperidis. 2005. Theoretical and Practical Issues in the Construction...
  • Simov, Kiril, Petya Osenova, Alexander Simov, and Milen Kouylekov. 2005. Design and implementation of the Bulgarian HPSG-based treebank. Journal...
  • van der Beek, Leonoor, Gosse Bouma, Robert Malouf, and Gertjan van Noord. 2002. The Alpino dependency treebank. In Computational Linguistics...