Joint Word and Morpheme Segmentation with Bayesian Non-Parametric Models - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Joint Word and Morpheme Segmentation with Bayesian Non-Parametric Models

Modèles bayésiens non-paramétriques pour la segmentation conjointe en mots et morphèmes


Language documentation often requires segmenting transcriptions of utterances collected on the field into words and morphemes. While these two tasks are typically performed in succession, we study here Bayesian models for simultaneously segmenting utterances at these two levels. Our aim is twofold: (a) to study the effect of explicitly introducing a hierarchy of units in joint segmentation models; (b) to further assess whether these two levels can be better identified through weak supervision. For this, we first consider a deterministic coupling between independent models; then design and evaluate hierarchical Bayesian models. Experiments with two under-resourced languages (Japhug and Tsez) allow us to better understand the value of various types of weak supervision. In our analysis, we use these results to revisit the distributional hypotheses behind Bayesian segmentation models and evaluate their validity for language documentation data.
Fichier principal
Vignette du fichier
2023.findings-eacl.48.pdf (639.31 Ko) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-04086368 , version 1 (02-05-2023)



  • HAL Id : hal-04086368 , version 1


Shu Okabe, François Yvon. Joint Word and Morpheme Segmentation with Bayesian Non-Parametric Models. 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2023), Association for Computational Linguistics, May 2023, Dubrovnik, Croatia. pp.628-642. ⟨hal-04086368⟩
190 Consultations
64 Téléchargements


Gmail Mastodon Facebook X LinkedIn More