home edit page issue tracker

This page pertains to UD version 2.

UD Icelandic Modern

Language: Icelandic (code: is)
Family: Indo-European, Germanic

This treebank has been part of Universal Dependencies since the UD v2.8 release.

The following people have contributed to making this treebank part of UD: Kristján Rúnarsson, Þórunn Arnardóttir, Hinrik Hafsteinsson, Starkaður Barkarson, Hildur Jónsdóttir, Steinþór Steingrímsson, Einar Freyr Sigurðsson.

Repository: UD_Icelandic-Modern
Search this treebank on-line: PML-TQ
Download all treebanks: UD 2.13

License: CC BY-SA 4.0

Genre: nonfiction, news

Questions, comments? General annotation questions (either Icelandic-specific or cross-linguistic) can be raised in the main UD issue tracker. You can report bugs in this treebank in the treebank-specific issue tracker on Github. If you want to collaborate, please contact [krunars (æt) gmail • com, thar (æt) hi • is, hinrik • hafst (æt) gmail • com, starkadur • barkarson (æt) arnastofnun • is, hildur • jonsdottir (æt) gmail • com, steinthor • steingrimsson (æt) arnastofnun • is, einar • freyr • sigurdsson (æt) arnastofnun • is]. Development of the treebank happens outside the UD repository. If there are bugs, either the original data source or the conversion procedure must be fixed. Do not submit pull requests against the UD repository.

Annotation Source
Lemmas annotated manually in non-UD style, automatically converted to UD
UPOS annotated manually in non-UD style, automatically converted to UD
XPOS annotated manually
Features annotated manually in non-UD style, automatically converted to UD
Relations annotated manually in non-UD style, automatically converted to UD

Description

UD_Icelandic-Modern is a conversion of the modern additions to the Icelandic Parsed Historical Corpus (IcePaHC) to the Universal Dependencies scheme.

The conversion was done using UDConverter.

Note that the treebank data in versions 2.8-2.10 is flawed in that various sentences are duplicated, resulting in an overlap between train / dev / test. This issue has been fixed as of version 2.11.

The modern additions to the Icelandic Parsed Historical Corpus (IcePaHC) is a corpus which includes 21st-century texts NOT found in IcePaHC. These texts were originally manually parsed according to the Penn Parsed Corpora of Historical English (PPCHE) annotation scheme as used for IcePaHC. These parsed texts were then automatically converted to the Universal Dependencies scheme to create UD_Icelandic-Modern. The corpus consists of 80,395 tokens in total. The texts are unprepared parliament speeches by four Icelandic members of parliament in the period 2011–2015 and sports news texts from two reporters at the Icelandic National Broadcasting Service (RÚV) from 2016 and – as of version 2.11 – from 2017. These 21st-century additions are not found in IcePaHC (which, however, contains two narrative texts from the 21st century).

Each sentence ID (sent_id) in UD_Icelandic-Modern carries the following information:

ALTHINGI_BO_2015_G-33-4647803,.1

Along with sentence IDs within the UD corpus, each sentence contains an X_ID flag, indicating the original sentence(s) in IcePaHC. This is useful as some sentences in the UD scheme have more than one original sentence in IcePaHC. As sentence IDs are currently not represented in the modern-text IcePaHC, this flag is left blank, as ID_MISSING for each original sentence.

Manually corrected portion

A subset of the UD corpus has been manually corrected after the automatic conversion, 15,151 words in total. The portion corrected belonged to the following files, as shown in each sentence’s ID:

Data split

Training data

61,817 words in total

Development data

8,431 words in total

Testing data

10,147 words in total

Acknowledgments

This project was funded by The Strategic Research and Development Programme for Language Technology, grant no. 180020-5301.

The modern additions to the Icelandic Parsed Historical Corpus (IcePaHC) are available here.

Morphological features were generated using ABLTagger, a PoS tagger for Icelandic, developed by Steinþór Steingrímsson, Örvar Kárason and Hrafn Loftsson and available here.

Statistics of UD Icelandic Modern

POS Tags

ADJADPADVAUXCCONJDETINTJNOUNNUMPARTPRONPROPNPUNCTSCONJVERBX

Features

CaseDefiniteDegreeForeignGenderMoodNumberNumTypePersonPronTypeTenseVerbFormVoice

Relations

aclacl:relcladvcladvmodamodapposauxcaseccccompcompoundcompound:prtconjcopcsubjdepdetdiscourseexplfixedflat:foreignflat:nameiobjmarknmodnmod:possnsubjnummodobjoblparataxispunctrootvocativexcomp

Tokenization and Word Segmentation

Morphology

Tags

Nominal Features

Degree and Polarity

Verbal Features

Pronouns, Determiners, Quantifiers

Other Features

Syntax

Auxiliary Verbs and Copula

Core Arguments, Oblique Arguments and Adjuncts

Here we consider only relations between verbs (parent) and nouns or pronouns (child).

Relations Overview