Word embeddings CLARIN.SI-embed.hr 1.0

Word embeddings CLARIN.SI-embed.hr 1.0

CLARIN.SI data & tools

Authors: Ljubešić, Nikola

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

Date issued: 2018-12-10

Type: lexicalConceptualResource, text

Size: 3147352 entries

Language(s): Croatian

Description: 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").

Publisher: Jožef Stefan Institute

Subject(s): word embeddings lemmatisation part-of-speech tagging

Collection(s): CLARIN.SI data & tools

This item is replaced by a newer submission:

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

Show full item record

Files in this item

This item is

Publicly Available

and licensed under:
Creative Commons - Attribution 4.0 International (CC BY 4.0)

Name: embed.hr-lemma.ft.sg.bin
Size: 1.82 GB
Format: Unknown
Description: Lemma embeddings, binary
MD5: 4308551842e5d3330d447d2896a0383e

Name: embed.hr-lemma.ft.sg.vec.gz
Size: 452.27 MB
Format: application/gzip
Description: Lemma embeddings, text format
MD5: 0cf64492d8adc3448b5c92740fe0854d

Name: embed.hr-token.ft.sg.bin
Size: 2.07 GB
Format: Unknown
Description: Token embeddings, binary
MD5: f892416d871a2607810f83ce3ffe2e20

Name: embed.hr-token.ft.sg.vec.gz
Size: 561.41 MB
Format: application/gzip
Description: Token embeddings, text format
MD5: 21b828b6287daa181f6eeb572a99948f