scholarly journals Terminology Translation in Low-Resource Scenarios

Information ◽  
2019 ◽  
Vol 10 (9) ◽  
pp. 273
Author(s):  
Rejwanul Haque ◽  
Mohammed Hasanuzzaman ◽  
Andy Way
Keyword(s):  
Machine Translation ◽  
End Users ◽  
Training Data ◽  
Classification Task ◽  
Domain Experts ◽  
Low Resource ◽  
Translation Quality ◽  
Classification Framework ◽  
Industrial Setting ◽  
Language Pair

Term translation quality in machine translation (MT), which is usually measured by domain experts, is a time-consuming and expensive task. In fact, this is unimaginable in an industrial setting where customised MT systems often need to be updated for many reasons (e.g., availability of new training data, leading MT techniques). To the best of our knowledge, as of yet, there is no publicly-available solution to evaluate terminology translation in MT automatically. Hence, there is a genuine need to have a faster and less-expensive solution to this problem, which could help end-users to identify term translation problems in MT instantly. This study presents a faster and less expensive strategy for evaluating terminology translation in MT. High correlations of our evaluation results with human judgements demonstrate the effectiveness of the proposed solution. The paper also introduces a classification framework, TermCat, that can automatically classify term translation-related errors and expose specific problems in relation to terminology translation in MT. We carried out our experiments with a low resource language pair, English–Hindi, and found that our classifier, whose accuracy varies across the translation directions, error classes, the morphological nature of the languages, and MT models, generally performs competently in the terminology translation classification task.

scholarly journals Augmenting Neural Machine Translation through Round-Trip Training Approach

2019 ◽  
Vol 9 (1) ◽  
pp. 268-278 ◽  
Author(s):  
Benyamin Ahmadnia ◽  
Bonnie J. Dorr
Keyword(s):  
Machine Translation ◽  
Training Data ◽  
Training Dataset ◽  
Round Trip ◽  
Neural Machine Translation ◽  
Low Resource ◽  
Translation Quality ◽  
High Resource ◽  
Training Approach ◽  
Language Pair

AbstractThe quality of Neural Machine Translation (NMT), as a data-driven approach, massively depends on quantity, quality and relevance of the training dataset. Such approaches have achieved promising results for bilingually high-resource scenarios but are inadequate for low-resource conditions. Generally, the NMT systems learn from millions of words from bilingual training dataset. However, human labeling process is very costly and time consuming. In this paper, we describe a round-trip training approach to bilingual low-resource NMT that takes advantage of monolingual datasets to address training data bottleneck, thus augmenting translation quality. We conduct detailed experiments on English-Spanish as a high-resource language pair as well as Persian-Spanish as a low-resource language pair. Experimental results show that this competitive approach outperforms the baseline systems and improves translation quality.

Improving thai-lao neural machine translation with similarity lexicon

2021 ◽  
pp. 1-10
Author(s):  
Zhiqiang Yu ◽  
Yuxin Huang ◽  
Junjun Guo
Keyword(s):  
Machine Translation ◽  
Semantic Information ◽  
Neural Machine Translation ◽  
Low Resource ◽  
Translation Quality ◽  
Decoder Architecture ◽  
Baseline System ◽  
Input Sentence ◽  
Resource Conditions ◽  
Language Pair

It has been shown that the performance of neural machine translation (NMT) drops starkly in low-resource conditions. Thai-Lao is a typical low-resource language pair of tiny parallel corpus, leading to suboptimal NMT performance on it. However, Thai and Lao have considerable similarities in linguistic morphology and have bilingual lexicon which is relatively easy to obtain. To use this feature, we first build a bilingual similarity lexicon composed of pairs of similar words. Then we propose a novel NMT architecture to leverage the similarity between Thai and Lao. Specifically, besides the prevailing sentence encoder, we introduce an extra similarity lexicon encoder into the conventional encoder-decoder architecture, by which the semantic information carried by the similarity lexicon can be represented. We further provide a simple mechanism in the decoder to balance the information representations delivered from the input sentence and the similarity lexicon. Our approach can fully exploit linguistic similarity carried by the similarity lexicon to improve translation quality. Experimental results demonstrate that our approach achieves significant improvements over the state-of-the-art Transformer baseline system and previous similar works.

scholarly journals Recurrent Stacking of Layers for Compact Neural Machine Translation Models

2019 ◽  
Vol 33 ◽  
pp. 6292-6299 ◽  
Author(s):  
Raj Dabre ◽  
Atsushi Fujita
Keyword(s):  
Machine Translation ◽  
Single Layer ◽  
Training Data ◽  
Neural Machine Translation ◽  
Parallel Corpora ◽  
Translation Quality ◽  
Sequence Generation ◽  
Sequence Modeling ◽  
Back Translation

In encoder-decoder based sequence-to-sequence modeling, the most common practice is to stack a number of recurrent, convolutional, or feed-forward layers in the encoder and decoder. While the addition of each new layer improves the sequence generation quality, this also leads to a significant increase in the number of parameters. In this paper, we propose to share parameters across all layers thereby leading to a recurrently stacked sequence-to-sequence model. We report on an extensive case study on neural machine translation (NMT) using our proposed method, experimenting with a variety of datasets. We empirically show that the translation quality of a model that recurrently stacks a single-layer 6 times, despite its significantly fewer parameters, approaches that of a model that stacks 6 different layers. We also show how our method can benefit from a prevalent way for improving NMT, i.e., extending training data with pseudo-parallel corpora generated by back-translation. We then analyze the effects of recurrently stacked layers by visualizing the attentions of models that use recurrently stacked layers and models that do not. Finally, we explore the limits of parameter sharing where we share even the parameters between the encoder and decoder in addition to recurrent stacking of layers.

scholarly journals Pseudotext Injection and Advance Filtering of Low-Resource Corpus for Neural Machine Translation

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Michael Adjeisah ◽  
Guohua Liu ◽  
Douglas Omwenga Nyabuga ◽  
Richard Nuetey Nortey ◽  
Jinling Song
Keyword(s):  
Machine Translation ◽  
Language Processing ◽  
Training Data ◽  
Target Language ◽  
Similarity Metrics ◽  
Mahalanobis Distances ◽  
Parallel Corpora ◽  
Parallel Corpus ◽  
Low Resource ◽  
Sentence Level

Scaling natural language processing (NLP) to low-resourced languages to improve machine translation (MT) performance remains enigmatic. This research contributes to the domain on a low-resource English-Twi translation based on filtered synthetic-parallel corpora. It is often perplexing to learn and understand what a good-quality corpus looks like in low-resource conditions, mainly where the target corpus is the only sample text of the parallel language. To improve the MT performance in such low-resource language pairs, we propose to expand the training data by injecting synthetic-parallel corpus obtained by translating a monolingual corpus from the target language based on bootstrapping with different parameter settings. Furthermore, we performed unsupervised measurements on each sentence pair engaging squared Mahalanobis distances, a filtering technique that predicts sentence parallelism. Additionally, we extensively use three different sentence-level similarity metrics after round-trip translation. Experimental results on a diverse amount of available parallel corpus demonstrate that injecting pseudoparallel corpus and extensive filtering with sentence-level similarity metrics significantly improves the original out-of-the-box MT systems for low-resource language pairs. Compared with existing improvements on the same original framework under the same structure, our approach exhibits tremendous developments in BLEU and TER scores.

scholarly journals Pre-Trained Transformer-Based Language Models for Sundanese

2021 ◽  
Author(s):  
Wilson Wongso ◽  
Henry Lucky ◽  
Derwin Suhartono
Keyword(s):  
Natural Language ◽  
Text Classification ◽  
Training Data ◽  
Language Models ◽  
Classification Task ◽  
Language Understanding ◽  
Training Corpus ◽  
Low Resource ◽  
Corpus Size ◽  
Fine Tune

Abstract The Sundanese language has over 32 million speakers worldwide, but the language has reaped little to no benefits from the recent advances in natural language understanding. Like other low-resource languages, the only alternative is to fine-tune existing multilingual models. In this paper, we pre-trained three monolingual Transformer-based language models on Sundanese data. When evaluated on a downstream text classification task, we found that most of our monolingual models outperformed larger multilingual models despite the smaller overall pre-training data. In the subsequent analyses, our models benefited strongly from the Sundanese pre-training corpus size and do not exhibit socially biased behavior. We released our models for other researchers and practitioners to use.

scholarly journals UQuAD1.0: Development of an Urdu Question Answering Training Data for Machine Reading Comprehension

2021 ◽  
Author(s):  
Samreen Ahmed ◽  
shakeel khoja
Keyword(s):  
Reading Comprehension ◽  
Machine Translation ◽  
Large Scale ◽  
Question Answering ◽  
Training Data ◽  
Significant Progress ◽  
Rule Based ◽  
Low Resource ◽  
Machine Reading ◽  
Answer Format

<p>In recent years, low-resource Machine Reading Comprehension (MRC) has made significant progress, with models getting remarkable performance on various language datasets. However, none of these models have been customized for the Urdu language. This work explores the semi-automated creation of the Urdu Question Answering Dataset (UQuAD1.0) by combining machine-translated SQuAD with human-generated samples derived from Wikipedia articles and Urdu RC worksheets from Cambridge O-level books. UQuAD1.0 is a large-scale Urdu dataset intended for extractive machine reading comprehension tasks consisting of 49k question Answers pairs in question, passage, and answer format. In UQuAD1.0, 45000 pairs of QA were generated by machine translation of the original SQuAD1.0 and approximately 4000 pairs via crowdsourcing. In this study, we used two types of MRC models: rule-based baseline and advanced Transformer-based models. However, we have discovered that the latter outperforms the others; thus, we have decided to concentrate solely on Transformer-based architectures. Using XLMRoBERTa and multi-lingual BERT, we acquire an F<sub>1</sub> score of 0.66 and 0.63, respectively.</p>

Neural machine translation of low-resource languages using SMT phrase pair injection

2020 ◽  
pp. 1-22
Author(s):  
Sukanta Sen ◽  
Mohammed Hasanuzzaman ◽  
Asif Ekbal ◽  
Pushpak Bhattacharyya ◽  
Andy Way
Keyword(s):  
Machine Translation ◽  
Large Scale ◽  
Production Systems ◽  
Statistical Machine Translation ◽  
Training Data ◽  
Original Training ◽  
Neural Machine Translation ◽  
Parallel Corpus ◽  
Low Resource ◽  
Better Than

Abstract Neural machine translation (NMT) has recently shown promising results on publicly available benchmark datasets and is being rapidly adopted in various production systems. However, it requires high-quality large-scale parallel corpus, and it is not always possible to have sufficiently large corpus as it requires time, money, and professionals. Hence, many existing large-scale parallel corpus are limited to the specific languages and domains. In this paper, we propose an effective approach to improve an NMT system in low-resource scenario without using any additional data. Our approach aims at augmenting the original training data by means of parallel phrases extracted from the original training data itself using a statistical machine translation (SMT) system. Our proposed approach is based on the gated recurrent unit (GRU) and transformer networks. We choose the Hindi–English, Hindi–Bengali datasets for Health, Tourism, and Judicial (only for Hindi–English) domains. We train our NMT models for 10 translation directions, each using only 5–23k parallel sentences. Experiments show the improvements in the range of 1.38–15.36 BiLingual Evaluation Understudy points over the baseline systems. Experiments show that transformer models perform better than GRU models in low-resource scenarios. In addition to that, we also find that our proposed method outperforms SMT—which is known to work better than the neural models in low-resource scenarios—for some translation directions. In order to further show the effectiveness of our proposed model, we also employ our approach to another interesting NMT task, for example, old-to-modern English translation, using a tiny parallel corpus of only 2.7K sentences. For this task, we use publicly available old-modern English text which is approximately 1000 years old. Evaluation for this task shows significant improvement over the baseline NMT.

POSTEDITING IN MACHINE TRANSLATION

2020 ◽  
Vol 2020 (30) ◽  
pp. 102-119
Author(s):  
Tetiana Korolova ◽  
Natalya Zhmayeva ◽  
Yulia Kolchah
Keyword(s):  
Machine Translation ◽  
Grammatical Structure ◽  
Incorrect Choice ◽  
Neural Machine Translation ◽  
Translation Quality ◽  
Modern Methods ◽  
Translation Services ◽  
Linguistic Units ◽  
Source Message ◽  
Language Pair

Modern industry of translation services singles out two translation quality levels that can be reached as a result of machine translation (MT) post-editing: good enough quality foresees rendering the main information of the source message, admitting stylistic, syntactic and morphological flaws while quality similar or equal to human translation is a full dress version of a post-edited text, ready to be published. The overview of MT systems enables us to consider Google Neural Machine Translation (GNMT) which is based on the most modern methods of training to reach maximum improvements the most powerful one. When analyzing texts translated by means of Google Translate the following problems were identified: distortion of the referential meaning of the source message, incorrect choice of variant equivalences, lack of terms harmonization, lack of abbreviations rendering, inconformity of linguistic units in persons, numbers and cases, incorrect choice of functional correspondings when rendering absolute constructions, gerund and participial constructions, literal translation of phrases, lack of transformations of the grammatical structure of the source message (additions, rearrangements). Taking into account the classified issues of machine translation as well as the levels of post-editing quality post-editing of the texts translated by means of MT is carried out, demands and recommendations applicable to post-editing results of MT within the language pair under analysis with respect to peculiarities of the specific MT system and the type of translated texts are provided.

Assessment of Multi-Engine Machine Translation for English to Hindi Language (MEMTEHiL)

2016 ◽  
Vol 6 (1) ◽  
pp. 30-45
Author(s):  
Pankaj K. Goswami ◽  
Sanjay K. Dwivedi ◽  
C. K. Jha
Keyword(s):  
Computer Science ◽  
Machine Translation ◽  
Translation Quality ◽  
Science Domain ◽  
Hindi Language ◽  
Interactive Version ◽  
Language Pair

English to Hindi translation of the computer-science related e-content, generated through an online freely available machine translation engine may not be technically correct. The expected target translation should be as fluent as intended for the native learners and the meaning of a source e-content should be conveyed properly. A Multi-Engine Machine Translation for English to Hindi Language (MEMTEHiL) framework has been designed and integrated by the authors as a translation solution for the computer science domain e-content. It was possible by enabling the use of well-tested approaches of machine translation. The humanly evaluated and acceptable metrics like fluency and adequacy (F&A) were used to assess the best translation quality for English to Hindi language pair. Besides humanly-judged metrics, another well-tested and existing interactive version of Bi-Lingual Evaluation Understudy (iBLEU) was used for evaluation. Authors have incorporated both parameters (F&A and iBLEU) for assessing the quality of translation as regenerated by the designed MEMTEHiL.

Sign in / Sign up

Export Citation Format

Share Document