Phonotactics as an Aid in Low Resource Loan Word Detection and Morphological Analysis in Sakha

Petter Mahlum, Sardana Ivanova

Forskningsoutput: Kapitel i bok/rapport/konferenshandlingKonferensbidragVetenskapligPeer review


Obtaining information about loan words and irregular morphological patterns can be difficult for low-resource languages. Using Sakha as an example, we show that it is possible to exploit known phonemic regularities such as vowel harmony and consonant distributions to identify loan words and irregular patterns, which can be helpful in rule-based downstream tasks such as parsing and POS-tagging. We evaluate phonemically inspired methods for loanword detection, combined with bigram vowel transition probabilities to inspect irregularities in the morphology of loanwords. We show that both these techniques can be useful for the detection of such patterns. Finally, we inspect the plural suffix -πAp [-LAr] to observe some of the variation in morphology between native and foreign words.

Titel på värdpublikationProceedings of the Second Workshop on Resources and Representations for Under-Resourced Languages and Domains (RESOURCEFUL-2023)
RedaktörerNikolai Ilinykh, Felix Morger, Dana Dannells, Simon Dobnik, Beata Megyesi, Joakim Nivre
Antal sidor10
FörlagThe Association for Computational Linguistics
ISBN (elektroniskt)978-1-959429-73-9
StatusPublicerad - 2023
MoE-publikationstypA4 Artikel i en konferenspublikation
EvenemangWorkshop on Resources and Representations for Under-Resourced Languages and Domains - Torshavn, Färöarna
Varaktighet: 22 maj 202322 maj 2023
Konferensnummer: 2

Bibliografisk information

Publisher Copyright:
© 2023 Association for Computational Linguistics.


  • 113 Data- och informationsvetenskap

Citera det här