Summary

Universal Derivations version of manually created Russian compounding relations extracted from Golden Compound Analyses (hhttp://nabil.hathout.free.fr/DeriMo2021/pdf-files/DeriMo_2021_paper_10.pdf).

Introduction

Golden Compound Analyses resource is a data set that capture Russian compounding relations which originates from Russkaya grammatika by Shvedova et al.

As the resource contains only compound lexemes, the data structure of the original resource was converted to the UDer format only (without necessity of additional harmonisation of the relations).

Acknowledgments

We wish to thank all the developers and annotators of the Golden Compound Analyses, including Daniil Vodolazsky, and Hermann Petrov.

References

As a citation for the resource in articles, please use this:

Vodolazsky, Daniil; Petrov, Hermann. 2021. Compound Splitting and Analysis for Russian. In Proceedings of the Third International Workshop on Resources and Tools for Derivational Morphology (DeriMo 2021). http://nabil.hathout.free.fr/DeriMo2021/pdf-files/DeriMo_2021_paper_10.pdf

@INPROCEEDINGS{GoldenCompoundAnalyses,
    title       = {{Compound Splitting and Analysis for Russian}},
    author      = {Vodolazsky, Daniil and Petrov, Hermann},
    booktitle   = {{Proceedings of the Third International Workshop on Resources and Tools for Derivational Morphology (DeriMo 2021)}},
    pages       = {145--153},
    year        = {2021},
    address     = {France},
    url         = {http://nabil.hathout.free.fr/DeriMo2021/pdf-files/DeriMo_2021_paper_10.pdf}
}

License

The resource is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC CC BY-NC 4.0). License text is available in the file LICENSE.txt.

=== Machine-readable metadata =================================================
Resource: Golden Compound Analyses
Language: Russian
Authors: Vodolazsky, Daniil; Petrov, Hermann
License: CC BY-NC 4.0
Contact: https://github.com/s231644/rucompoundsplitter/tree/main/data/gold_analyses
===============================================================================

=== Machine-readable metadata =================================================
Harmonized resource: Golden Compound Analyses
Harmonized version: 1.1
Data source: https://github.com/s231644/rucompoundsplitter/tree/main/data/gold_analyses
Data available since: UDer v1.1
Harmonization: default
Common features: Morphological categories; Compounding
JSON features: dataset
Lemmas: 4931
Relations: 1639
Families: 3292
Singletons: 2348
Avarage tree size: 1.5
Avarage tree depth: 0.3
Avarage tree out-degree: 0.5
Maximum tree size: 24
Maximum tree depth: 2
Maximum tree out-degree: 24
Part-of-speech: ADJ, 21.4; ADV, 1.7; Affixoid, 0.9; NOUN, 47.6; NUM, 1.8; PART, 1.1; PRON, 1.7; PROPN, 0.4; VERB, 11.8; X, 11.5
Derivational relations: 0
Conversion relations: 0
Compounding relations: 1639
Variant relations: 0
===============================================================================