Slovak Dependency Treebank

  Authors

Gajdošová, Katarína ; Šimková, Mária and et al.

  Item identifier

http://hdl.handle.net/11234/1-1822

 Project URL

http://korpus.juls.savba.sk/

 Date issued

2016-11-07

 Type

corpus, text

 Size

10604 sentences, 106043 tokens

 Language(s)

Slovak

 Description

Slovak Dependency Treebank (Slovenský závislostný korpus) was created as part of the Slovak National Corpus at the Ľ. Štúr Institute of the Slovak Academy of Sciences. The annotation follows the guidelines of the Prague Dependency Treebank (Czech), slightly modified in the spirit of Slovak grammatical tradition. Morphological tags, lemmas and dependency relations have been assigned manually to every word. The present dataset is a subset of the original treebank. We automatically selected the sentences where the two human annotators 100% agreed on the analysis. This increases the quality and trustworthiness of the data but it also results in selecting short sentences most of the time. An extended version may be published in the future when manually merged and checked annotation is available. The selected sentences have been converted to the CoNLL-X file format (original token IDs are preserved in the FEATS column). This PDT-style annotation will serve as the source for the first Slovak dataset in the Universal Dependencies (to be published separately).

 Publisher

Jazykovedný ústav Ľ. Štúra Slovenskej akadémie vied

 Acknowledgement

Ministerstvo školstva SR

Project code: 2003SP200280307

Project name: Komplexné spracovanie slovenského jazyka a jeho elektronizácia na účely jazykovedného výskumu

 

Grantová agentura České republiky

Project code: 15-10472S

Project name: Morphologically and Syntactically Annotated Corpora of Many Languages

 Subject(s)

dependency treebank syntax morphology

 Collection(s)

LINDAT / CLARIAH-CZ Data & Tools

Show full item record