Advantages of Dependency Parsing for Free Word Order Natural Languages

  • Seyed Amin Mirlohi Falavarjani
  • Gholamreza Ghassem-Sani
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8939)


An important reason to prefer dependency parsing over classical phrased based methods, especially for languages such as Persian, with the property of being “free word order”, is that this particular property has a negative impact on the accuracy of conventional parsing methods. In Persian, some words such as adverbs can freely be moved within a sentence without affecting its correctness or meaning. In this paper, we illustrate the robustness of dependency parsing against this particular problem by training two well-known dependency parsers, namely MST Parser and Malt Parser, using a Persian dependency corpus called Dadegan. We divided the corpus into two separate parts including only projective sentences and only non-projective sentences, which are corelated with the free word order property. As our results show, MST Parsing is not only more tolerant than Malt Parsing against the free word order problem, but it is also in general a more accurate technique.


data mining knowledge discovery and machine learning knowledge modeling and processing 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Agirre, E., Kepa, B., Koldo, G., Nivre, J.: Improving dependency parsing with semantic classes. In: Association for Computational Linguistics: Human Language Technologies: short papers, pp 699–703 (2011)Google Scholar
  2. 2.
    Cortes, C., Vapnik, V.: Support-Vector Networks. Machine Learning, pp. 273–297 (1995)Google Scholar
  3. 3.
    Heshaam, F., Ghassem-Sani, G.: Unsupervised grammar induction using history based approach. In: Computer Speech & Language, pp. 644–658 (2006)Google Scholar
  4. 4.
    Klein, D., Manning, C.D.: Natural language grammar induction using a constituent-context model. In: Advances in Neural Information Processing Systems, pp. 35–42 (2001)Google Scholar
  5. 5.
    Klein, D., Manning, C.D.: Corpus-based induction of syntactic structure: Models of dependency and constituency. In: Association for Computational Linguistics (2004)Google Scholar
  6. 6.
    Koo, T., Carreras, X., Collins, M.: Simple semi-supervised dependency parsing. In: ACL, pp. 595–603 (2008)Google Scholar
  7. 7.
    McDonald, R., Pereira, F., Ribarov, K., Hajic, J.: Non-projective dependency parsing using spanning tree algorithms. In: Human Language Technology and Empirical Methods in Natural Language Processing (HLT/EMNLP), pp. 523–530 (2005)Google Scholar
  8. 8.
    Mesgar, M., Ghasem-Sani, G.: History Based Unsupervised Data Oriented Parsing. In: RANLP, pp. 453–459 (2013)Google Scholar
  9. 9.
    Mirroshandel, S.A., Ghassem-Sani, G.: Unsupervised Grammar Induction Using a Parent Based Constituent Context Model. In: ECAI, pp. 293–297 (2008)Google Scholar
  10. 10.
    Nivre, J.: Dependency grammar and dependency parsing. Technical Report MSI (2005)Google Scholar
  11. 11.
    Nivre, J., Hall, J., Nilsson, J.: Maltparser: A data-driven parser-generator for dependency parsing. In: Language Resources and Evaluation (LREC), Genoa, Italy, pp. 2216–2219 (2006)Google Scholar
  12. 12.
    Rasooli, M.S., Kouhestani, M., Moloodi, A.: Development of a Persian Syntactic Dependency Treebank. In: North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 306–314 (2013)Google Scholar
  13. 13.
    Seraji, M., Beatam, M., Nivre, J.: Dependency parsers for Persian. In: Asian Language Resources, COLING, pp. 35–43 (2012)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Seyed Amin Mirlohi Falavarjani
    • 1
  • Gholamreza Ghassem-Sani
    • 1
  1. 1.Department of Computer EngineeringSharif University of technologyTehranIran

Personalised recommendations