Modelling Verbal Morphology in Nen

Saliha Muradoglu, Nicholas Evans, Ekaterina Vylomova


Abstract
Nen verbal morphology is particularly complex; a transitive verb can take up to 1,740 unique forms. The combined effect of having a large combinatoric space and a low-resource setting amplifies the need for NLP tools. Nen morphology utilises distributed exponence - a non-trivial means of mapping form to meaning. In this paper, we attempt to model Nen verbal morphology using state-of-the-art machine learning models for morphological reinflection. We explore and categorise the types of errors these systems generate. Our results show sensitivity to training data composition; different distributions of verb type yield different accuracies (patterning with E-complexity). We also demonstrate the types of patterns that can be inferred from the training data, through the case study of sycretism.
Anthology ID:
2020.alta-1.5
Volume:
Proceedings of the 18th Annual Workshop of the Australasian Language Technology Association
Month:
December
Year:
2020
Address:
Virtual Workshop
Editors:
Maria Kim, Daniel Beck, Meladel Mistica
Venue:
ALTA
SIG:
Publisher:
Australasian Language Technology Association
Note:
Pages:
43–53
Language:
URL:
https://aclanthology.org/2020.alta-1.5
DOI:
Bibkey:
Cite (ACL):
Saliha Muradoglu, Nicholas Evans, and Ekaterina Vylomova. 2020. Modelling Verbal Morphology in Nen. In Proceedings of the 18th Annual Workshop of the Australasian Language Technology Association, pages 43–53, Virtual Workshop. Australasian Language Technology Association.
Cite (Informal):
Modelling Verbal Morphology in Nen (Muradoglu et al., ALTA 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.alta-1.5.pdf