Filters








598 Hits in 6.1 sec

Multi-word Expressions in English-Latvian Machine Translation

Inguna Skadiņa
2016 Baltic Journal of Modern Computing  
The paper presents series of experiments that aim to find best method how to treat multi-word expressions (MWE) in machine translation task.  ...  Methods have been investigated in a framework of statistical machine translation (SMT) for translation form English into Latvian.  ...  Acknowledgements The research was supported by Grant 271/2012 from the Latvian Council of Science. This work has been supported by the IC1207 COST Action PARSEME as part of its scientific program.  ... 
doi:10.22364/bjmc.2016.4.4.14 fatcat:bszbtth2j5hsle56paq5reyfhi

Paying Attention to Multi-Word Expressions in Neural Machine Translation [article]

Matīss Rikters, Ondřej Bojar
2019 arXiv   pre-print
Processing of multi-word expressions (MWEs) is a known problem for any natural language processing task. Even neural machine translation (NMT) struggles to overcome it.  ...  This paper presents results of experiments on investigating NMT attention allocation to the MWEs and improving automated translation of sentences that contain MWEs in English->Latvian and English->Czech  ...  Acknowledgement This study was supported in parts by the grants H2020-ICT-2014-1-645442 (QT21), the ICT COST Action IC1207 ParseME: Parsing and multi-word expressions.  ... 
arXiv:1710.06313v2 fatcat:uifhjiaaanhx3fpqqqxybfl4ai

English-Latvian Toponym Processing: Translation Strategies and Linguistic Patterns

Tatiana Gornostay, Inguna Skadina
2009 European Association for Machine Translation Conferences/Workshops  
investigated and implemented for English-Latvian machine translation. 10,000 The UK-related toponyms from Geonames were selected for a development set.  ...  The evaluation of output quality on basis of a test set has showed 67% accuracy in out-ofvocabulary translation: 58% on a set containing one-word toponymic units and 81% on a multi-word test set.  ...  LTTPs can be of two types: in-word patterns and multi-word patterns.  ... 
dblp:conf/eamt/GornostayS09 fatcat:bn3ozf42jjhwdevt5mnom5hosy

Sentiment Analysis in Latvian and Russian: A Survey

Rinalds Vīksna, Gints Jēkabsons
2018 Applied Computer Systems  
Most studies are carried out on the basis of texts written in English, while other languages are being less researched.  ...  The present survey focuses on research conducted on the sentiment analysis for the Latvian and Russian languages.  ...  [13] Machine translation and existing sentiment analysis tools for English Latvian translated to English Twitter (multi- domain) Positive, (Neutral), Negative Accuracy 76 % without neutral  ... 
doi:10.2478/acss-2018-0006 fatcat:perlu6a2hvg3rjgnmcfoxqzs4a

Hybrid Machine Translation by Combining Output from Multiple Machine Translation Systems

Matīss Rikters
2019 Baltic Journal of Modern Computing  
of resulting translations; MT combination systems for combining output from EnglishLatvian statistical MT.  ...  The key results of this research are new state-of-the-art machine translation systems for English ↔ Estonian; approaches for utilising neural MT generated attention alignments for MT combination and comprehension  ...  Multi-word expressions (MWEs) have been a challenge for SMT and NMT because they may not appear frequently enough in training data.  ... 
doi:10.22364/bjmc.2019.7.3.01 fatcat:z6uyzz3wxzfwrhon7h3oeumefa

Grammatical Error Correction and Style Transfer via Zero-shot Monolingual Translation [article]

Elizaveta Korotkova, Agnes Luhtaru, Maksym Del, Krista Liin, Daiga Deksne, Mark Fishel
2019 arXiv   pre-print
The model is trained on bilingual data in all translation directions (English-to-Estonian, Estonian-to-English, English-to-Latvian, etc.)  ...  ("Ma tahan" / "I want" in Estonian, "two" and "saldējumus" / "ice-creams" in genitive, plural in Latvian) is correctly translated into English as "I want two ice creams.".  ... 
arXiv:1903.11283v2 fatcat:5mny42lqsfaihan7ml32abzi5e

Contextual Encoding for Translation Quality Estimation

Junjie Hu, Wei-Cheng Chang, Yuexin Wu, Graham Neubig
2018 Proceedings of the Third Conference on Machine Translation: Shared Task Papers  
The task of word-level quality estimation (QE) consists of taking a source sentence and machine-generated translation, and predicting which words in the output are correct and which are wrong.  ...  The first part uses an embedding layer to represent words and their part-of-speech tags in both languages.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official poli-cies, either expressed or implied, of the U.S.  ... 
doi:10.18653/v1/w18-6462 dblp:conf/wmt/HuCWN18 fatcat:wds566ri6rcd3pecnojpdzconm

The QT21 Combined Machine Translation System for English to Latvian

Jan-Thorsten Peter, Hermann Ney, Ondřej Bojar, Ngoc-Quan Pham, Jan Niehues, Alex Waibel, Franck Burlot, François Yvon, Mārcis Pinnis, Valters Sics, Joost Bastings, Miguel Rios (+4 others)
2017 Proceedings of the Second Conference on Machine Translation  
This paper describes the joint submission of the QT21 projects for the EnglishLatvian translation task of the EMNLP 2017 Second Conference on Machine Translation (WMT 2017).  ...  The submission is a system combination which combines seven different statistical machine translation systems provided by the different groups.  ...  Members of the QT21 project have jointly built a combined statistical machine translation system, in order to achieve high-quality machine translation from English into Latvian.  ... 
doi:10.18653/v1/w17-4734 dblp:conf/wmt/PeterNBPNWBYPSB17 fatcat:vdolqpulcjertofnaphqjui43y

MT Detection in Web-Scraped Parallel Corpora

Spencer Rarrick, Chris Quirk, William Lewis
2011 Machine Translation Summit  
In this paper, we present an algorithm for filtering machine-translated content from Webscraped parallel corpora, and discuss its application in cleaning such corpora for use in training statistical machine  ...  The Web is an invaluable source of parallel data, but in recent years it has become polluted with increasing amounts of machine-translated content.  ...  , and English-German, and General for English-Latvian).  ... 
dblp:conf/mtsummit/RarrickQL11 fatcat:i34sm4cs7zfj7otgs7xeh7cl5a

Finding Translation Examples for Under-Resourced Language Pairs or for Narrow Domains; the Case for Machine Translation

Dan Tufis
2012 Computer Science Journal of Moldova  
The cyberspace is populated with valuable information sources, expressed in about 1500 different languages and dialects.  ...  Recent advancements in cross-lingual information retrieval, multilingual summarization, cross-lingual question answering and machine translation promise to narrow the linguistic gaps and lower the communication  ...  On-line machine translation services such as those offered by Google, Microsoft or Yahoo allow for assimilation of knowledge originally expressed in tens of languages unknown to the standard reader.  ... 
doaj:05a53f2716f24a9b87820f9505c4adb6 fatcat:nymorko3vze73gyaukjk7sou64

Contextual Encoding for Translation Quality Estimation [article]

Junjie Hu, Wei-Cheng Chang, Yuexin Wu, Graham Neubig
2018 arXiv   pre-print
The task of word-level quality estimation (QE) consists of taking a source sentence and machine-generated translation, and predicting which words in the output are correct and which are wrong.  ...  The first part uses an embedding layer to represent words and their part-of-speech tags in both languages.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official poli-cies, either expressed or implied, of the U.S.  ... 
arXiv:1809.00129v1 fatcat:irk6jawiezb2hj7rf7vhue5pfe

MultiMWE: Building a Multi-lingual Multi-Word Expression (MWE) Parallel Corpora [article]

Lifeng Han, Gareth J.F. Jones, Alan F. Smeaton
2020 arXiv   pre-print
MWEs in other NLP fields such as Machine Translation.  ...  Multi-word expressions (MWEs) are a hot topic in research in natural language processing (NLP), including topics such as MWE detection, MWE decomposition, and research investigating the exploitation of  ...  English to Latvian MT.  ... 
arXiv:2005.10583v1 fatcat:dqf4ujgozjhglfzwhmofa4mlde

Evaluation of contextual embeddings on less-resourced languages [article]

Matej Ulčar and Aleš Žagar and Carlos S. Armendariz and Andraž Repar and Senja Pollak and Matthew Purver and Marko Robnik-Šikonja
2021 arXiv   pre-print
Most existing work focuses on English; in contrast, we present here the first multilingual empirical comparison of two ELMo and several monolingual and multilingual BERT models using 14 tasks in nine languages  ...  The current dominance of deep neural networks in natural language processing is based on contextual embeddings such as ELMo, BERT, and BERT derivatives.  ...  Partial support was also received from the UK EP-SRC under grant EP/S033564/1, and the EPSRC/AHRC Centre for Doctoral Training in Media and Arts Technology EP/L01632X/1.  ... 
arXiv:2107.10614v1 fatcat:vhbn3v2xqzgx5gzysnmb3jtq3e

Application of Online Terminology Services in Statistical Machine Translation

Raivis Skadins, Marcis Pinnis, Tatiana Gornostay, Andrejs Vasiljevs
2013 Machine Translation Summit  
statistical machine translation during training and translation phases.  ...  We focus on the use case for the application of online terminology services in statistical machine translation and describe the applied methods for monolingual and bilingual terminology integration into  ...  A recent study in the FP7 project TTC (2013) has shown that for English-Latvian the preprocessing does not yield positive results for term translation.  ... 
dblp:conf/mtsummit/SkadinsPGV13 fatcat:mtzy42grszfoln5k4vhpk2o3wq

The WMT'18 Morpheval test suites for English-Czech, English-German, English-Finnish and Turkish-English

Franck Burlot, Yves Scherrer, Vinit Ravishankar, Ondřej Bojar, Stig-Arne Grönroos, Maarit Koponen, Tommi Nieminen, François Yvon
2018 Proceedings of the Third Conference on Machine Translation: Shared Task Papers  
In this paper, we extend the Morpheval protocol introduced by Burlot and Yvon (2017) for the English-to-Czech and English-to-Latvian translation directions to three additional language pairs, and report  ...  Progress in the quality of machine translation output calls for new automatic evaluation procedures and metrics.  ...  Research and Innovation Programme under Grant Agreement No 780069 and by the grant 18-24210S of the Czech Science Foundation. 14 Unfortunately, we did not obtain the output of the Alibaba-Ensemble system in  ... 
doi:10.18653/v1/w18-6433 dblp:conf/wmt/BurlotSRBGKNY18 fatcat:5prdnwk4f5bz3l73xcvaghjzbu
« Previous Showing results 1 — 15 out of 598 results