Croatian Dependency Treebank: Recent Development and Initial Experiments

We present the current state of development of the Croatian Dependency Treebank – with special empahsis on adapting the Prague Dependency Treebank formalism to Croatian language specifics – and illustrate its possible applications in an experiment with dependency parsing using MaltParser. The treeba...

Full description

Permalink: http://skupnikatalog.nsk.hr/Record/ffzg.KOHA-OAI-FFZG:317846/Details
Matična publikacija: Proceedings of the Eigth International Conference on Language Resources and Evaluation (LREC'12)
Istanbul : European Language Resources Association (ELRA), 2012
Glavni autori: Berović, Daša (-), Tadić, Marko (Author), Agić, Željko
Vrsta građe: Članak
Jezik: eng
Online pristup: http://bib.irb.hr/datoteka/580881.719_Paper.pdf
http://www.lrec-conf.org/proceedings/lrec2012/index.html
LEADER 02581naa a2200289uu 4500
008 131111s2012 xx 1 eng|d
035 |a (CROSBI)580881 
040 |a HR-ZaFF  |b hrv  |c HR-ZaFF  |e ppiak 
100 1 |9 839  |a Berović, Daša 
245 1 0 |a Croatian Dependency Treebank: Recent Development and Initial Experiments /  |c Berović, Daša ; Agić, Željko ; Tadić, Marko. 
246 3 |i Naslov na engleskom:  |a Croatian Dependency Treebank: Recent Development and Initial Experiments 
300 |a 1902-1906  |f str. 
520 |a We present the current state of development of the Croatian Dependency Treebank – with special empahsis on adapting the Prague Dependency Treebank formalism to Croatian language specifics – and illustrate its possible applications in an experiment with dependency parsing using MaltParser. The treebank currently contains approximately 2870 sentences, out of which the 2699 sentences and 66930 tokens were used in this experiment. Three linear-time projective algorithms implemented by the MaltParser system – Nivre eager, Nivre standard and stack projective – running on default settings were used in the experiment. The highest performing system, implementing the Nivre eager algorithm, scored (LAS 71.31 UAS 80.93 LA 83.87) within our experiment setup. The results obtained serve as an illustration of treebank’s usefulness in natural language processing research and as a baseline for further research in dependency parsing of Croatian. 
536 |a Projekt MZOS  |f 130-1300646-0645 
536 |a Projekt MZOS  |f 130-1300646-1776 
546 |a ENG 
690 |a 5.04 
690 |a 6.03 
693 |a dependency treebank, dependency parsing, Croatian language  |l hrv  |2 crosbi 
693 |a dependency treebank, dependency parsing, Croatian language  |l eng  |2 crosbi 
700 1 |9 888  |a Tadić, Marko  |4 aut 
700 1 |9 495  |a Agić, Željko  |4 aut 
773 0 |a Eigth International Conference on Language Resources and Evaluation (LREC'12) (23-25.05.2012. ; Istanbul, Turska)  |t Proceedings of the Eigth International Conference on Language Resources and Evaluation (LREC'12)  |d Istanbul : European Language Resources Association (ELRA), 2012  |n Calzolari, Nicoletta ; Choukri, Khalid ; Declerck, Thierry ; Ugur Dogan, Mehmet ; Maegaard, Bente ; Mariani, Joseph ; Odijk, Jan ; Piperidis, Stelios  |z 978-2-9517408-7-7  |g str. 1902-1906 
856 |u http://bib.irb.hr/datoteka/580881.719_Paper.pdf 
856 |u http://www.lrec-conf.org/proceedings/lrec2012/index.html 
942 |c RZB  |u 2  |v MeđRecenzija  |z Znanstveni - Poster - CijeliRad  |t 1.08 
999 |c 317846  |d 317844