NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH
Neural machine translation is a method used in automatic translation that makes use of artificial neural networks. A single model takes an input sequence and predicts the most likely output sequence of words after being trained on parallel data. In this master thesis, a neural machine translation mo...
Main Author: | |
---|---|
Other Authors: | , |
Format: | Text |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/2077/70914 |
id |
ftunivgoeteborg:oai:gupea.ub.gu.se:2077/70914 |
---|---|
record_format |
openpolar |
spelling |
ftunivgoeteborg:oai:gupea.ub.gu.se:2077/70914 2023-10-29T02:38:49+01:00 NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH Pfau, Merle University of Gothenburg / Department of Philosophy,Lingustics and Theory of Science Göteborgs universitet / Institutionen för filosofi, lingvistik och vetenskapsteori 2022-03-08T08:07:06Z application/pdf https://hdl.handle.net/2077/70914 eng eng https://hdl.handle.net/2077/70914 Neural Machine Translation low-resource language North Sámi - Swedish Text H2 Student essay 2022 ftunivgoeteborg 2023-10-04T21:18:09Z Neural machine translation is a method used in automatic translation that makes use of artificial neural networks. A single model takes an input sequence and predicts the most likely output sequence of words after being trained on parallel data. In this master thesis, a neural machine translation model for the language pair North Sámi - Swedish was developed and trained. Since no parallel corpus exists between the two languages, a data set of Norwegian and North Sámi of about 225.000 sentences was translated to Swedish and used as training data. The model architecture is based on Vaswani et al. (2017)’s transformer, which is the state-of-the-art approach, if enough parallel data is available. Following Sennrich et al. (2016)’s techniques of combining methods to lower the amount of necessary data, a BLEU score of 44.11 was achieved. Due to the relatively small amount of available parallel data, techniques of incorporating monolingual bitext and creating synthetic additional data were implemented, but did not result in any further improvements. Text North Sámi Sámi University of Gothenburg: GUPEA (Gothenburg University Publications Electronic Archive) |
institution |
Open Polar |
collection |
University of Gothenburg: GUPEA (Gothenburg University Publications Electronic Archive) |
op_collection_id |
ftunivgoeteborg |
language |
English |
topic |
Neural Machine Translation low-resource language North Sámi - Swedish |
spellingShingle |
Neural Machine Translation low-resource language North Sámi - Swedish Pfau, Merle NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
topic_facet |
Neural Machine Translation low-resource language North Sámi - Swedish |
description |
Neural machine translation is a method used in automatic translation that makes use of artificial neural networks. A single model takes an input sequence and predicts the most likely output sequence of words after being trained on parallel data. In this master thesis, a neural machine translation model for the language pair North Sámi - Swedish was developed and trained. Since no parallel corpus exists between the two languages, a data set of Norwegian and North Sámi of about 225.000 sentences was translated to Swedish and used as training data. The model architecture is based on Vaswani et al. (2017)’s transformer, which is the state-of-the-art approach, if enough parallel data is available. Following Sennrich et al. (2016)’s techniques of combining methods to lower the amount of necessary data, a BLEU score of 44.11 was achieved. Due to the relatively small amount of available parallel data, techniques of incorporating monolingual bitext and creating synthetic additional data were implemented, but did not result in any further improvements. |
author2 |
University of Gothenburg / Department of Philosophy,Lingustics and Theory of Science Göteborgs universitet / Institutionen för filosofi, lingvistik och vetenskapsteori |
format |
Text |
author |
Pfau, Merle |
author_facet |
Pfau, Merle |
author_sort |
Pfau, Merle |
title |
NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
title_short |
NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
title_full |
NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
title_fullStr |
NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
title_full_unstemmed |
NEURAL MACHINE TRANSLATION FROM NORTH SÁMI TO SWEDISH |
title_sort |
neural machine translation from north sámi to swedish |
publishDate |
2022 |
url |
https://hdl.handle.net/2077/70914 |
genre |
North Sámi Sámi |
genre_facet |
North Sámi Sámi |
op_relation |
https://hdl.handle.net/2077/70914 |
_version_ |
1781065181427662848 |