Reproducing Czech Syntactic Parsing Results Published in CoNLL Tasks

Logo poskytovatele

Varování

Publikace nespadá pod Ekonomicko-správní fakultu, ale pod Fakultu informatiky. Oficiální stránka publikace je na webu muni.cz.
Autoři

KOCINCOVÁ Lucia

Rok publikování 2012
Druh Článek ve sborníku
Konference 6th Workshop on Recent Advances in Slavonic Natural Language Processing
Fakulta / Pracoviště MU

Fakulta informatiky

Citace
www https://nlp.fi.muni.cz/raslan/2012/paper12.pdf
Obor Informatika
Klíčová slova syntactic analysis; parsing; Czech; parser evaluation
Popis In this paper, I describe the approach on reproducing MST Parser and MaltParser results for Czech dependency parsing which were published in several tasks of Conference on Computational Natural Language Learning. Firstly, I briefly describe basic principles of parsers. Then, I include features and options that need to be optimized while using the parsers to get the desired results on testing files as well as evaluation methodology used in my research. I also shortly mention hardware requirements for those, who would like to train their own model for Czech language parsing. Finally, I include the practical application of trained models for our approach.
Související projekty:

Používáte starou verzi internetového prohlížeče. Doporučujeme aktualizovat Váš prohlížeč na nejnovější verzi.