UD Estonian is a converted version of the Estonian Dependency Treebank (EDT), originally annotated in the Constraint Grammar (CG) annotation scheme, and consisting of genres of fiction, newspaper texts and scientific texts. The treebank contains 30,972 trees, 437,769 tokens.
The Estonian UD EDT treebank is based on the Estonian Dependency Treebank (EDT), created at the University of Tartu. The treebank has been automatically converted and then manually reviewed and reannotated.
The treebank covers 3 different genres, namely newspaper texts, fiction and scientific texts:
- fiction (67,744 tokens, 5,522 sentences)
- newspapers (266,806 tokens, 18,690 sentences)
- scientific texts (94,022 tokens, 5,483 sentences)
- Also, the subpart of Estonian part of HamleDT 3.0 treebank has been reannotated and included in the treebank; it contains 9,200 tokens in 1277 sentences.
In addition to standard ud annotation, annotation of named entiites has been added to MISC-field (NE=B-Type or NE=I-Type, there TYPE stands for PER (person), ORG (organisation), LOC (location), GEP (geopolitical name), EVENT (events), PROD (product), MUU (other) or UNK (unknown)).
Also, a preliminary argument structure has been annotated in the MISC field (Verb=meaning, Arg=verb_Arg_[0-5]).
We wish to thank all who have contributed to the original EDT annotation effort, especially Eleri Aedmaa, Riin Kirt and Dage Särg.
We also thank developers of udapi, ud annotatrix and , and ConlluEditor tools.
This work was financed by the National Programme for Estonian Language Technology and Estonian Ministery of Education and Research (grant 20-56 IUT20-56 "Computational models for Estonian").
- Kadri Muischnek, Kaili Müürisep, Tiina Puolakainen, Eleri Aedmaa, Riin Kirt, Dage Särg. 2014. Estonian Dependency Treebank and its annotation scheme. In: Proceedings of the 13th Workshop on Treebanks and Linguistic Theories (TLT13), pp. 285–291, ISBN 978-3-9809183-9-8, Tübingen, Germany.
- Kadri Muischnek, Kaili Müürisep and Tiina Puolakainen 2016. Estonian Dependency Treebank: from Constraint Grammar tagset to Universal Dependencies. - Proceedings of LREC 2016.
- Kadri Muischnek and Kaili Müürisep. 2017. Estonian copular and existential constructions as an UD annotation problem. In Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017), pp. 79-85. 2017.
- Kadri Muischnek, Kaili Müürisep. 2023. Named Entity layer in Estonian UD treebanks. In Proceedings NoDaLiDa 2023.
- UD v2.17: minor errors fixed, mostly related to obl:agent dependencies.
- UD v2.16: ExtPos feature added, annotation of the argument structure improved.
- UD v2.15: annotation of foreign phrases improved, fixed errors in annotation of participles and determiners, agent adverbials have special tag obl:agent, added special layer of argument structure of frequent verbs to the misc field (this annotation is preliminary, annotated verbs have the feature Verb=lemma or Verb=lemma_digit (if the verb has many very different meanings and argument patterns), arguments have the feature Arg=verblemma_Arg_digit).
- UD v2.13: annotation of enhanced dependencies improved
- UD v2.12: minor errors fixed
- UD v2.11: annotation of enhanced dependencies improved; fixed issues of reported speech, numbers; added named entity annotation to MISC field (NE=B-Type or NE=I-Type where Type stands for Per-person, Loc - location, Gep - geopolitical institution, Org - organisation, Eve - event, Prod - product, Muu - other, Unk - unknown, B stands for beginning and I stands for in).
- UD v2.10: annotation of enhanced dependencies improved, subjects of clausal complements (xcomp) are annotated, regardless of whether this subject is the same or different from the subject of the next higher clause; errors of goeswith annotation fixed, annotation of foreign words improved.
- UD v2.8: Morphological annotation of numerals has been improved, also better documentation has been included
- UD v2.6: In the enhanced representation, 0-nodes have been added in clauses in which a predicate is elided, and the relative pronoun has been attached to its antecedent with the relation 'ref'.
- UD v2.5: Various individual annotation errors and inconsistencies solved. 3527 tokens added to training data.
- UD v2.4: xcomp function rechecked, some fixed constructions added. Various individual annotatation errors and inconsistencies solved.
- UD v2.3: more data added to v2.2. Now the whole original EDT is included in Estonian UD. Annotation of elliptical constructions (label 'orphan') is more systematic. Various individual annotation errors and inconsistencies solved.
- UD v2.2: more data added to v2.1; fixed errors in v2.1 files; repository renamed from UD_Estonian to UD_Estonian-EDT.
- UD v2.1: manual reannotation of copula sentences, names and appositions; semiautomatic reannotation of pronouns and determiners and coordinated structures; automatic reannotation of nmod and obl functions.
- UD v2.0: manual reannotation of copula sentences, names and appositions; semiautomatic reannotation of pronouns and determiners and coordinated structures; automatic reannotation of nmod and obl functions.
- UD v1.2 contained Arborest, a much smaller and older VISL-style treebank. It has been re-annotated and added to EDT for UD v1.3.
=== Machine-readable metadata ================================================= Documentation status: stub Data source: semi-automatic Data available since: UD v1.2 License: CC BY-NC-SA 4.0 Includes text: yes Parallel: no Genre: fiction news nonfiction academic Lemmas: converted from manual UPOS: converted from manual XPOS: converted from manual Features: converted from manual Relations: converted from manual Contributing: here Contributors: Muischnek, Kadri; Müürisep, Kaili; Puolakainen, Tiina; Rääbis, Andriela; Torga, Liisi Contact: [email protected], [email protected] ===============================================================================