1 / 22

SynCro - Parsing Simple Croatian Sentences

SynCro - Parsing Simple Croatian Sentences. Kristina Vučković, Božo Bekavac , Zdravko Dovedan University of Zagreb, Faculty of Humanities and Social Science { kvuckovi, bbekavac ,zdovedan } @ ffzg.hr NooJ2009 Tozeur 2009-06- 09. Overview of the work.

emma
Download Presentation

SynCro - Parsing Simple Croatian Sentences

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. SynCro - Parsing Simple Croatian Sentences Kristina Vučković, Božo Bekavac, Zdravko Dovedan University of Zagreb, Faculty of Humanities and Social Science {kvuckovi, bbekavac,zdovedan}@ffzg.hr NooJ2009Tozeur2009-06-09 NooJ2009Tozeur 2009-06-09

  2. Overview of the work • on our way to complete parse trees for Croatian sentences • building partial trees for simple Croatian sentences • favor of precision over recall • Building syntactic parse trees for Croatian is quite a challenge! • properties of Croatian language: • mostly free word (better to say constituent?) order → scrambling very frequent • could be a long distance between parts of predicate NooJ2009Tozeur 2009-06-09

  3. Syntactic trees • at this stage, the following structures of parse trees are described • verb • subject - verb • subject - verb - direct object • subject - verb - direct (and, or indirect object) • in addition to all these combinations, we can detect • adverbial phrases of place and • adverbial phrases of time (temporal expressions) • Complex combinations (e.g. multiple PPs) are not described NooJ2009Tozeur 2009-06-09

  4. Sentance structure in Croatian <subject> <predicate> <predicate> <object_D> <object_ID> <A_time> <subject> <A_place> <predicate> <subject> NooJ2009Tozeur 2009-06-09

  5. Recognition of: predicate node NooJ2009Tozeur 2009-06-09

  6. Recognition of: subject node NooJ2009Tozeur 2009-06-09

  7. Recognition of: direct object node NooJ2009Tozeur 2009-06-09

  8. Recognition of: indirect object node NooJ2009Tozeur 2009-06-09

  9. Recognition of: Adverbs of Time NooJ2009Tozeur 2009-06-09

  10. Recognition of: Adverbs of Place NooJ2009Tozeur 2009-06-09

  11. Exampless NooJ2009Tozeur 2009-06-09

  12. Exampless NooJ2009Tozeur 2009-06-09

  13. Exampless NooJ2009Tozeur 2009-06-09

  14. Exampless NooJ2009Tozeur 2009-06-09

  15. Exampless NooJ2009Tozeur 2009-06-09

  16. Exampless • dislocated predicate <PRED> NooJ2009Tozeur 2009-06-09

  17. Exampless • dislocated predicate <PRED> NooJ2009Tozeur 2009-06-09

  18. Exampless • dislocated predicate <PRED> NooJ2009Tozeur 2009-06-09

  19. Exampless • dislocated predicate <PRED> and dislocated subject <SUB> NooJ2009Tozeur 2009-06-09

  20. Addition to the Lexicon • For recognizing Adverbials of Time and Place: • Adjectives +time / + place • R+vr • R+mt • Prepositions +time / + place • S+vr • S+mt NooJ2009Tozeur 2009-06-09

  21. if there is a nominal predicate if the <NP> node is a coordination made of 2 or more <NP>’s of different gender if <S> has more than one annotation ex. The child was very smart. -> The child was ->was very smart. Problems and Future Work • ex. • The boy and a girl were very smart. ->The boy and a girl =><NP+p+m> • ex. • He was singing from March to April. • He was singing from Paris to Tozeur. NooJ2009Tozeur 2009-06-09

  22. NooJ2009Tozeur 2009-06-09

More Related