home edit page issue tracker

This page pertains to UD version 2.

UD Finnish FTB

Language: Finnish (code: fi)
Family: Uralic, Finnic

This treebank has been part of Universal Dependencies since the UD v1.1 release.

The following people have contributed to making this treebank part of UD: Jussi Piitulainen, Hanna Nurmi.

Repository: UD_Finnish-FTB
Search this treebank on-line: PML-TQ
Download all treebanks: UD 2.2

License: CC BY 4.0

Genre: grammar-examples

Questions, comments? General annotation questions (either Finnish-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 [fin-clarin (æt) helsinki • fi]. 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


FinnTreeBank 1 consists of manually annotated grammatical examples from VISK. The UD version of FinnTreeBank 1 was converted from a native annotation model with a script.

The UD version of FinnTreeBank 1 was derived from FinnTreeBank 1 2014 by a scripted mapping of labels and some restructuring in an attempt to conform approximately to the UD Finnish model.

Unannotated linguistic material in FinnTreeBank 1 2014 was adapted from the examples in Ison suomen kieliopin verkkoversio (The Web Version of the Large Grammar of Finnish, VISK), available on-line as http://scripta.kotus.fi/visk, and annotations originally produced and further revised in FIN-CLARIN projects in the Department of Modern Languages, University of Helsinki.

Differences from the UD Finnish model

Some differences from the UD Finnish model remain unaddressed.

Surface words that have an adverb or a conjunction fused with a following negative verb are separated into two tokens in the FinnTreeBank model. The UD version keeps them as such but adds the unannotated joined token before the two.

Most punctuation tokens are linked to a nearby token instead of a clause head.

As a catchall, the dep relation is used as intended when a more proper mapping could not be determined.

Some FinnTreeBank annotations are retained in the MISC field.


The previous version of the treebank was split into training, development, and test sets by repeatedly taking 8 sentences into training set, 1 into development set, and 1 into test set.


Statistics of UD Finnish FTB

POS Tags






Tokenization and Word Segmentation



Nominal Features

Degree and Polarity

Verbal Features

Pronouns, Determiners, Quantifiers

Other Features


Auxiliary Verbs and Copula

Core Arguments, Oblique Arguments and Adjuncts

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

Verbs with Reflexive Core Objects

Relations Overview