Abu-MaTran at WMT 2016 Translation Task: Deep Learning,
            Morphological Segmentation and Tuning on Character Sequences

Sánchez-Cartagena, Víctor M.; Toral, Antonio

doi:10.18653/v1/w16-2322

Cited by 14 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural MT: NMT is a new approach to statistical MT, which, while having been introduced only very recently, 2 has already shown great potential, as there is evidence that it can attain better translation quality than the dominant approach to date, namely phrase-based statistical MT (PBSMT). This has been shown for a number of language pairs and domains, including transcribed speeches (Luong and Manning, 2015), newswire (Sánchez-Cartagena and Toral, 2016) and United Nations documents (Junczys-Dowmunt et al, 2016). Beyond its generally positive performance, NMT is of particular interest for literary texts due to the following two findings:…”

mentioning

confidence: 86%

What Level of Quality Can Neural Machine Translation Attain on Literary Text?

Toral

Way

2018

Machine Translation: Technologies and Applications

Self Cite

View full text Add to dashboard Cite

Given the rise of a new approach to MT, Neural MT (NMT), and its promising performance on different text types, we assess the translation quality it can attain on what is perceived to be the greatest challenge for MT: literary text. Specifically, we target novels, arguably the most popular type of literary text. We build a literary-adapted NMT system for the English-to-Catalan translation direction and evaluate it against a system pertaining to the previous dominant paradigm in MT: statistical phrase-based MT (PBSMT). To this end, for the first time we train MT systems, both NMT and PBSMT, on large amounts of literary text (over 100 million words) and evaluate them on a set of twelve widely known novels spanning from the the 1920s to the present day. According to the BLEU automatic evaluation metric, NMT is significantly better than PBSMT (p < 0.01) on all the novels considered. Overall, NMT results in a 11% relative improvement (3 points absolute) over PBSMT. A complementary human evaluation on three of the books shows that between 17% and 34% of the translations, depending on the book, produced by NMT (versus 8% and 20% with PBSMT) are perceived by native speakers of the target language to be of equivalent quality to translations produced by a professional human translator.

show abstract

mentioning

confidence: 86%

What Level of Quality Can Neural Machine Translation Attain on Literary Text?

Toral

Way

2018

Machine Translation: Technologies and Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…Bradbury and Socher (Bradbury et al, 2014) employed the modified Morfessor to provide morphology knowledge into word segmentation, but they neglected the morphological varieties between subword units, which might result in am- and Toral (Sánchez-Cartagena et al, 2016) proposed a rule-based morphological word segmentation for Finnish, which applies BPE on all the morpheme units uniformly without distinguishing their inner morphological roles. Huck (Huck et al, 2017) explored target-side segmentation method for German, which shows that the cascading of suffix splitting and compound splitting with BPE can achieve better translation results.…”

Section: Related Workmentioning

confidence: 99%

Morphological Word Segmentation on Agglutinative Languages for Neural Machine Translation

Pan,

Li,

Yang

et al. 2020

Preprint

View full text Add to dashboard Cite

Neural machine translation (NMT) has achieved impressive performance on machine translation task in recent years. However, in consideration of efficiency, a limited-size vocabulary that only contains the top-N highest frequency words are employed for model training, which leads to many rare and unknown words. It is rather difficult when translating from the lowresource and morphologically-rich agglutinative languages, which have complex morphology and large vocabulary. In this paper, we propose a morphological word segmentation method on the sourceside for NMT that incorporates morphology knowledge to preserve the linguistic and semantic information in the word structure while reducing the vocabulary size at training time. It can be utilized as a preprocessing tool to segment the words in agglutinative languages for other natural language processing (NLP) tasks. Experimental results show that our morphologically motivated word segmentation method is better suitable for the NMT model, which achieves significant improvements on Turkish-English and Uyghur-Chinese machine translation tasks on account of reducing data sparseness and language complexity. Sennrich, Rico and Haddow, Barry and Birch, Alexandra. Improving neural machine translation models with monolingual data. In Computer Science, 2016.

show abstract

A product and process analysis of post-editor corrections on neural, statistical and rule-based machine translation output

Koponen

Salmi

Nikulin

2019

Machine Translation

View full text Add to dashboard Cite

This paper presents a comparison of post-editing (PE) changes performed on English-to-Finnish neural (NMT), rule-based (RBMT) and statistical machine translation (SMT) output, combining a product-based and a process-based approach. A total of 33 translation students acted as participants in a PE experiment providing both post-edited texts and edit process data. Our product-based analysis of the postedited texts shows statistically significant differences in the distribution of edit types between machine translation systems. Deletions were the most common edit type for the RBMT, insertions for the SMT, and word form changes as well as word substitutions for the NMT system. The results also show significant differences in the correctness and necessity of the edits, particularly in the form of a large number of unnecessary edits in the RBMT output. Problems related to certain verb forms and ambiguity were observed for NMT and SMT, while RBMT was more likely to handle them correctly. Process-based comparison of effort indicators shows a slight increase of keystrokes per word for NMT output, and a slight decrease in average pause length for NMT compared to RBMT and SMT in specific text blocks. A statistically significant difference was observed in the number of visits per sub-segment, which is lower for NMT than for RBMT and SMT. The results suggest that although different types of edits were needed to outputs from NMT, RBMT and SMT systems, the difference is not necessarily reflected in process-based effort indicators.

show abstract

Abu-MaTran at WMT 2016 Translation Task: Deep Learning, Morphological Segmentation and Tuning on Character Sequences

Cited by 14 publications

References 17 publications

What Level of Quality Can Neural Machine Translation Attain on Literary Text?

What Level of Quality Can Neural Machine Translation Attain on Literary Text?

Morphological Word Segmentation on Agglutinative Languages for Neural Machine Translation

A product and process analysis of post-editor corrections on neural, statistical and rule-based machine translation output

Contact Info

Product

Resources

About