A case study demonstrating an approach to the statistical analysis of the variation of multiword expressions in Slovene corpora

Čibej, Jaka (author)

In Slovene linguistics, much research in phraseology has either been theoretical in nature or focused more on compiling lexicographic resources for human users. While several machine-readable lexicographic resources containing multiword expressions (MWEs) have also been developed in recent years, Slovene phraseology and computational Slovene linguistics remain largely divided into separate tracks. We attempt to bridge the gap with a brief demonstration of the benefits that computational and statistical approaches based on machine-readable data can have for linguists and phraseologists. We briefly present the SUK Training Corpus of Slovene, the largest machine-readable dataset for Slovene that contains annotations of multiword expressions, as well as the Q-CAT Corpus Annotation Tool that was used to annotate it. We extract examples for two Slovene MWEs (priti na zeleno vejo and podirati se kot hišica iz kart) from the morphosyntactically annotated Gigafida 2.1 Corpus of Written Standard Slovene using a rule-based approach that leverages syntactic structures. We perform a statistical analysis to determine the degree of variation within the extracted examples. We aim to show that machine-readable data is intended not only for developers of NLP tools but can also help provide additional insight into the structure and variation for the linguistic description of MWEs.

• Language: English
• Keywords: multiword expressions, multiword expression variants, statistical analysis, automatic extraction, corpora
• Work type: Other
• Typology: 1.08 - Published Scientific Conference Contribution
• Organization: FRI - Faculty of Computer and Information Science
• Publication status: Published
• Publication version: Version of Record
• Year: 2025
• Number of pages: Str. 45-61
• Numbering: Letn. 65, št. 1
• PID: 20.500.12556/RUL-177754
• UDC: 81'322
• ISSN on article: 0024-3922
• DOI: 10.4312/linguistica.65.1.45-61
• COBISS.SI-ID: 263491843
• Publication date in RUL: 06.01.2026

Record is a part of a proceedings

• Title: Frazeologija v digitalni dobi
• COBISS.SI-ID: 262399235

Record is a part of a journal

• Title: Linguistica
• Shortened title: Linguistica
• Publisher: Filozofska fakulteta, Filozofska fakulteta, Znanstvena založba Filozofske fakultete, Založba Univerze v Ljubljani
• ISSN: 0024-3922
• COBISS.SI-ID: 16075264

Licences

• License: CC BY-SA 4.0, Creative Commons Attribution-ShareAlike 4.0 International
• Link: http://creativecommons.org/licenses/by-sa/4.0/
• Description: This Creative Commons license is very similar to the regular Attribution license, but requires the release of all derivative works under this same license.

Secondary language

• Language: Slovenian
• Title: Študija primera za prikaz pristopa k statistični analizi variantnosti večbesednih enot v slovenskih korpusih

Abstract

V slovenskem jezikoslovju je precej raziskav na področju frazeologije teoretične na-rave ali pa so usmerjene v gradnjo slovarskih virov za človeške uporabnike. V zadnjih letih je bilo zgrajenih tudi nekaj strojno berljivih slovarskih virov, ki vsebujejo večbe-sedne enote (VBE), a slovenska frazeologija in slovensko računalniško jezikoslovje večinoma ostajata ločeni. V prispevku ju poskusimo povezati s kratkim prikazom ko-risti, ki jih lahko računalniški in statistični pristopi na podlagi strojno berljivih podat-kov prinesejo jezikoslovcem in frazeologom. Na kratko predstavimo učni korpus pisne slovenščine SUK, največjo strojno berljivo slovensko podatkovno množico z oznakami večbesednih enot, ter orodje za označevanje korpusov Q-CAT, s katerim je bil korpus označen. Iz oblikoskladenjsko označenega korpusa pisne standardne slovenščine Giga-fida 2.1 nato izluščimo primere za dve slovenski VBE (priti na zeleno vejo in podirati se kot hišica iz kart) s pomočjo pristopa na podlagi pravil, ki se opira na skladenjske strukture. Izvedemo statistično analizo in določimo stopnjo variantnosti znotraj izluš-čenih primerov. S tem želimo pokazati, da strojno berljivi podatki niso namenjeni le razvijalcem orodij za obdelavo naravnega jezika, temveč lahko tudi jezikoslovcem po-nudijo dodaten vpogled v strukturo in variantnost VBE.

• Keywords: večbesedne enote, variante večbesednih enot, statistična analiza, strojno luščenje, korpusi