Low-Resource Active Learning of North Sámi Morphological Segmentation
Many Uralic languages have a rich morphological structure, but lack tools of morphological analysis needed for efficient language processing. While creating a high-quality morphological analyzer requires a significant amount of expert labor, data-driven approaches may provide sufficient quality for...
Published in: | Septentrio Conference Series |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Septentrio Academic Publishing
2015
|
Subjects: | |
Online Access: | https://septentrio.uit.no/index.php/SCS/article/view/3465 https://doi.org/10.7557/5.3465 |
id |
ftunitroemsoe:oai:ojs.henry.ub.uit.no:article/3465 |
---|---|
record_format |
openpolar |
spelling |
ftunitroemsoe:oai:ojs.henry.ub.uit.no:article/3465 2023-05-15T17:40:07+02:00 Low-Resource Active Learning of North Sámi Morphological Segmentation Grönroos, Stig-Arne Jokinen, Kristiina Hiovain, Katri Kurimo, Mikko Virpioja, Sami 2015-06-17 application/pdf https://septentrio.uit.no/index.php/SCS/article/view/3465 https://doi.org/10.7557/5.3465 eng eng Septentrio Academic Publishing https://septentrio.uit.no/index.php/SCS/article/view/3465/3389 https://septentrio.uit.no/index.php/SCS/article/view/3465 doi:10.7557/5.3465 Copyright (c) 2015 Stig-Arne Grönroos, Kristiina Jokinen, Katri Hiovain, Mikko Kurimo, Sami Virpioja Septentrio Conference Series; No 2 (2015): First International Workshop on Computational Linguistics for Uralic Languages; 20–33 2387-3086 10.7557/scs.2015.2 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Peer-reviewed Article 2015 ftunitroemsoe https://doi.org/10.7557/5.3465 https://doi.org/10.7557/scs.2015.2 2021-08-16T16:49:15Z Many Uralic languages have a rich morphological structure, but lack tools of morphological analysis needed for efficient language processing. While creating a high-quality morphological analyzer requires a significant amount of expert labor, data-driven approaches may provide sufficient quality for many applications.We study how to create a statistical model for morphological segmentation of North Sámi language with a large unannotated corpus and a small amount of human-annotated word forms selected using an active learning approach. For statistical learning, we use the semi-supervised Morfessor Baseline and FlatCat methods. Aer annotating 237 words with our active learning setup, we improve morph boundary recall over 20% with no loss of precision. Article in Journal/Newspaper North Sámi Sámi University of Tromsø: Septentrio Academic Publishing Septentrio Conference Series 2 20 |
institution |
Open Polar |
collection |
University of Tromsø: Septentrio Academic Publishing |
op_collection_id |
ftunitroemsoe |
language |
English |
description |
Many Uralic languages have a rich morphological structure, but lack tools of morphological analysis needed for efficient language processing. While creating a high-quality morphological analyzer requires a significant amount of expert labor, data-driven approaches may provide sufficient quality for many applications.We study how to create a statistical model for morphological segmentation of North Sámi language with a large unannotated corpus and a small amount of human-annotated word forms selected using an active learning approach. For statistical learning, we use the semi-supervised Morfessor Baseline and FlatCat methods. Aer annotating 237 words with our active learning setup, we improve morph boundary recall over 20% with no loss of precision. |
format |
Article in Journal/Newspaper |
author |
Grönroos, Stig-Arne Jokinen, Kristiina Hiovain, Katri Kurimo, Mikko Virpioja, Sami |
spellingShingle |
Grönroos, Stig-Arne Jokinen, Kristiina Hiovain, Katri Kurimo, Mikko Virpioja, Sami Low-Resource Active Learning of North Sámi Morphological Segmentation |
author_facet |
Grönroos, Stig-Arne Jokinen, Kristiina Hiovain, Katri Kurimo, Mikko Virpioja, Sami |
author_sort |
Grönroos, Stig-Arne |
title |
Low-Resource Active Learning of North Sámi Morphological Segmentation |
title_short |
Low-Resource Active Learning of North Sámi Morphological Segmentation |
title_full |
Low-Resource Active Learning of North Sámi Morphological Segmentation |
title_fullStr |
Low-Resource Active Learning of North Sámi Morphological Segmentation |
title_full_unstemmed |
Low-Resource Active Learning of North Sámi Morphological Segmentation |
title_sort |
low-resource active learning of north sámi morphological segmentation |
publisher |
Septentrio Academic Publishing |
publishDate |
2015 |
url |
https://septentrio.uit.no/index.php/SCS/article/view/3465 https://doi.org/10.7557/5.3465 |
genre |
North Sámi Sámi |
genre_facet |
North Sámi Sámi |
op_source |
Septentrio Conference Series; No 2 (2015): First International Workshop on Computational Linguistics for Uralic Languages; 20–33 2387-3086 10.7557/scs.2015.2 |
op_relation |
https://septentrio.uit.no/index.php/SCS/article/view/3465/3389 https://septentrio.uit.no/index.php/SCS/article/view/3465 doi:10.7557/5.3465 |
op_rights |
Copyright (c) 2015 Stig-Arne Grönroos, Kristiina Jokinen, Katri Hiovain, Mikko Kurimo, Sami Virpioja |
op_doi |
https://doi.org/10.7557/5.3465 https://doi.org/10.7557/scs.2015.2 |
container_title |
Septentrio Conference Series |
container_issue |
2 |
container_start_page |
20 |
_version_ |
1766140925017325568 |