Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation

This article presents a probabilistic hierarchical clustering model for morphological segmentation. In contrast to existing approaches to morphology learning, our method allows learning hierarchical organization of word morphology as a collection of tree structured paradigms. The model is fully unsu...

Full description

Bibliographic Details
Main Authors: Burcu Can, Suresh Manandhar
Format: Article
Language:English
Published: The MIT Press 2018-06-01
Series:Computational Linguistics
Online Access:https://www.mitpressjournals.org/doi/pdf/10.1162/coli_a_00318
id doaj-175f8b45e5854742b8f1694bdd62f720
record_format Article
spelling doaj-175f8b45e5854742b8f1694bdd62f7202020-11-25T01:45:01ZengThe MIT PressComputational Linguistics1530-93122018-06-0144234937410.1162/coli_a_00318coli_a_00318Tree Structured Dirichlet Processes for Hierarchical Morphological SegmentationBurcu Can0Suresh Manandhar1Hacettepe University, Department of Computer Engineering. burcucan@cs.hacettepe.edu.trUniversity of York, Department of Computer Science. suresh@cs.york.ac.ukThis article presents a probabilistic hierarchical clustering model for morphological segmentation. In contrast to existing approaches to morphology learning, our method allows learning hierarchical organization of word morphology as a collection of tree structured paradigms. The model is fully unsupervised and based on the hierarchical Dirichlet process. Tree hierarchies are learned along with the corresponding morphological paradigms simultaneously. Our model is evaluated on Morpho Challenge and shows competitive performance when compared to state-of-the-art unsupervised morphological segmentation systems. Although we apply this model for morphological segmentation, the model itself can also be used for hierarchical clustering of other types of data.https://www.mitpressjournals.org/doi/pdf/10.1162/coli_a_00318
collection DOAJ
language English
format Article
sources DOAJ
author Burcu Can
Suresh Manandhar
spellingShingle Burcu Can
Suresh Manandhar
Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
Computational Linguistics
author_facet Burcu Can
Suresh Manandhar
author_sort Burcu Can
title Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
title_short Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
title_full Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
title_fullStr Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
title_full_unstemmed Tree Structured Dirichlet Processes for Hierarchical Morphological Segmentation
title_sort tree structured dirichlet processes for hierarchical morphological segmentation
publisher The MIT Press
series Computational Linguistics
issn 1530-9312
publishDate 2018-06-01
description This article presents a probabilistic hierarchical clustering model for morphological segmentation. In contrast to existing approaches to morphology learning, our method allows learning hierarchical organization of word morphology as a collection of tree structured paradigms. The model is fully unsupervised and based on the hierarchical Dirichlet process. Tree hierarchies are learned along with the corresponding morphological paradigms simultaneously. Our model is evaluated on Morpho Challenge and shows competitive performance when compared to state-of-the-art unsupervised morphological segmentation systems. Although we apply this model for morphological segmentation, the model itself can also be used for hierarchical clustering of other types of data.
url https://www.mitpressjournals.org/doi/pdf/10.1162/coli_a_00318
work_keys_str_mv AT burcucan treestructureddirichletprocessesforhierarchicalmorphologicalsegmentation
AT sureshmanandhar treestructureddirichletprocessesforhierarchicalmorphologicalsegmentation
_version_ 1725025775650865152