fdemelo's picture
Update README.md
d232057 verified
metadata
language:
  - ar
  - bg
  - bxr
  - ca
  - cs
  - cu
  - da
  - de
  - el
  - en
  - es
  - et
  - eu
  - fa
  - fi
  - fr
  - ga
  - gl
  - got
  - grc
  - he
  - hi
  - hr
  - hsb
  - hu
  - id
  - it
  - ja
  - kk
  - kmr
  - ko
  - la
  - lv
  - nb
  - nl
  - nn
  - pl
  - pt
  - pt
  - ro
  - ru
  - sk
  - sl
  - sme
  - sv
  - tr
  - ug
  - uk
  - ur
  - vi
  - zh
license: cc-by-nc-sa-4.0

UD CoNLL-U 2017 aligned

This dataset is a processed version of the dataset made available for the UD CoNLL Shared Task 2017, entitled "Multilingual Parsing from Raw Text to Universal Dependencies." This task is based on the Universal Dependencies 2.0 dataset.

The processing aligns words and tokens along with their morphological annotations using the universal part-of-speech set for sequence-to-sequence task training.

The dataset fields are described in the table below:

Field Description
id Universal Dependencies submission ID
lang IETF language code (cf. langcodes)
text Text sentence
tags Sequence of universal POS tags

Sequence of Part-of-Speech tags

The sequence of POS tags is separated by spaces, but contiguity of words with punctuation and contractions from combinations of words are marked by a + character. Example:

Text: You   won't     see   this section.
Tags: PRON  AUX+PART  VERB  DET  NOUN+PUNCT

Note that won't, as equivalent to will+not, is tagged as AUX+PART accordingly. Also, the contiguity between section and the period renders the tagging as NOUN+PUNCT.

Languages

The following languages are available in the dataset.

Tag Language
ar Arabic
bg Bulgarian
ca Catalan
cs Czech
cu Old Church Slavic
da Danish
de German
el Greek
en English
es Spanish
et Estonian
eu Basque
fa Persian
fi Finnish
fr French
ga Irish
gl Galician*
got Gothic
grc Ancient Greek
he Hebrew
hi Hindi
hr Croatian
hu Hungarian
id Indonesian
it Italian
ja Japanese
kk Kazakh*
ko Korean
la Latin
lv Latvian
nb Norwegian Bokmål
nl Dutch
nn Norwegian Nynorsk
pl Polish
pt Portuguese
pt-BR Portuguese (Brazil)
ro Romanian
ru Russian
sk Slovak
sl Slovenian
sv Swedish
tr Turkish
ug Uyghur*
uk Ukrainian*
ur Urdu
vi Vietnamese
zh Chinese

* Not available in the development subset.

Only in the test subset, the following "surprise" languages are added:

Tag Language
bxr Buriat
hsb Upper Sorbian
kmr Northern Kurdish
sme Northern Sami

License

The distribution license is inherited from the original dataset and so available to free copy and use under the CC BY-SA or CC BY-NC-SA licenses, depending on the language. See UD licenses for more details.