Word embeddings CLARIN.SI-embed.hr 1.0

  Avtorji

Ljubešić, Nikola

  Identifikator vnosa

http://hdl.handle.net/11356/1205

 Datum objave

2018-12-10

 Vrsta

lexicalConceptualResource, text

 Velikost

3147352 entries

 Jezik(i)

Croatian

 Opis

CLARIN.SI-embed.hr contains word embeddings induced from a large collection of Croatian texts composed of the Croatian web corpus hrWaC and a 400-million-token-heavy collection of newspaper texts. The embeddings are based on the skip-gram model of fastText trained on 1,852,631,924 tokens of running text for (1) 1,742,837 lowercased surface forms (e.g., "hrvatske") and (2) 1,404,515 lowercased lemmas with added part-of-speech information (e.g., "hrvatska#Np").

 Izdajatelj

Jožef Stefan Institute

 Zahvala

Jožef Stefan Institute CLARIN "CLARIN.SI"

 Ključne besede

word embeddings lemmatisation part-of-speech tagging

 Zbirke

CLARIN.SI data & tools

 

Ta vnos je bil nadomeščen z novejšim.

http://hdl.handle.net/11356/1790

Seznam vseh različic  

Prikaži polni zapis vnosa