Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map
Neural machine translation (NMT) methods based on various artificial neural network models have shown remarkable performance in diverse tasks and have become mainstream for machine translation currently. Despite the recent successes of NMT applications, a predefined vocabulary is still required, mea...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/11/15/7026 |
_version_ | 1797525770922360832 |
---|---|
author | Jangwon Lee Jungi Lee Minho Lee Gil-Jin Jang |
author_facet | Jangwon Lee Jungi Lee Minho Lee Gil-Jin Jang |
author_sort | Jangwon Lee |
collection | DOAJ |
description | Neural machine translation (NMT) methods based on various artificial neural network models have shown remarkable performance in diverse tasks and have become mainstream for machine translation currently. Despite the recent successes of NMT applications, a predefined vocabulary is still required, meaning that it cannot cope with out-of-vocabulary (OOV) or rarely occurring words. In this paper, we propose a postprocessing method for correcting machine translation outputs using a named entity recognition (NER) model to overcome the problem of OOV words in NMT tasks. We use attention alignment mapping (AAM) between the named entities of input and output sentences, and mistranslated named entities are corrected using word look-up tables. The proposed method corrects named entities only, so it does not require retraining of existing NMT models. We carried out translation experiments on a Chinese-to-Korean translation task for Korean historical documents, and the evaluation results demonstrated that the proposed method improved the bilingual evaluation understudy (BLEU) score by 3.70 from the baseline. |
first_indexed | 2024-03-10T09:18:42Z |
format | Article |
id | doaj.art-4e2842f8043a40978d632f4ca1debdaa |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-10T09:18:42Z |
publishDate | 2021-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-4e2842f8043a40978d632f4ca1debdaa2023-11-22T05:23:18ZengMDPI AGApplied Sciences2076-34172021-07-011115702610.3390/app11157026Named Entity Correction in Neural Machine Translation Using the Attention Alignment MapJangwon Lee0Jungi Lee 1Minho Lee 2Gil-Jin Jang3SK Holdings C&C, Gyeonggi-do, Suwon City 13558, KoreaDepartment of Artificial Intelligence, Kyungpook National University, Daegu 41566, KoreaDepartment of Artificial Intelligence, Kyungpook National University, Daegu 41566, KoreaSchool of Electronic and Electrical Engineering, Kyungpook National University, Daegu 41566, KoreaNeural machine translation (NMT) methods based on various artificial neural network models have shown remarkable performance in diverse tasks and have become mainstream for machine translation currently. Despite the recent successes of NMT applications, a predefined vocabulary is still required, meaning that it cannot cope with out-of-vocabulary (OOV) or rarely occurring words. In this paper, we propose a postprocessing method for correcting machine translation outputs using a named entity recognition (NER) model to overcome the problem of OOV words in NMT tasks. We use attention alignment mapping (AAM) between the named entities of input and output sentences, and mistranslated named entities are corrected using word look-up tables. The proposed method corrects named entities only, so it does not require retraining of existing NMT models. We carried out translation experiments on a Chinese-to-Korean translation task for Korean historical documents, and the evaluation results demonstrated that the proposed method improved the bilingual evaluation understudy (BLEU) score by 3.70 from the baseline.https://www.mdpi.com/2076-3417/11/15/7026neural networksrecurrent neural networksnatural language processingneural machine translationnamed entity recognition |
spellingShingle | Jangwon Lee Jungi Lee Minho Lee Gil-Jin Jang Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map Applied Sciences neural networks recurrent neural networks natural language processing neural machine translation named entity recognition |
title | Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map |
title_full | Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map |
title_fullStr | Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map |
title_full_unstemmed | Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map |
title_short | Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map |
title_sort | named entity correction in neural machine translation using the attention alignment map |
topic | neural networks recurrent neural networks natural language processing neural machine translation named entity recognition |
url | https://www.mdpi.com/2076-3417/11/15/7026 |
work_keys_str_mv | AT jangwonlee namedentitycorrectioninneuralmachinetranslationusingtheattentionalignmentmap AT jungilee namedentitycorrectioninneuralmachinetranslationusingtheattentionalignmentmap AT minholee namedentitycorrectioninneuralmachinetranslationusingtheattentionalignmentmap AT giljinjang namedentitycorrectioninneuralmachinetranslationusingtheattentionalignmentmap |