Matches in SemOpenAlex for { <https://semopenalex.org/work/W4300692318> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4300692318 abstract "The necessity of using a fixed-size word vocabulary in order to control the model complexity in state-of-the-art neural machine translation (NMT) systems is an important bottleneck on performance, especially for morphologically rich languages. Conventional methods that aim to overcome this problem by using sub-word or character-level representations solely rely on statistics and disregard the linguistic properties of words, which leads to interruptions in the word structure and causes semantic and syntactic losses. In this paper, we propose a new vocabulary reduction method for NMT, which can reduce the vocabulary of a given input corpus at any rate while also considering the morphological properties of the language. Our method is based on unsupervised morphology learning and can be, in principle, used for pre-processing any language pair. We also present an alternative word segmentation method based on supervised morphological analysis, which aids us in measuring the accuracy of our model. We evaluate our method in Turkish-to-English NMT task where the input language is morphologically rich and agglutinative. We analyze different representation methods in terms of translation accuracy as well as the semantic and syntactic properties of the generated output. Our method obtains a significant improvement of 2.3 BLEU points over the conventional vocabulary reduction technique, showing that it can provide better accuracy in open vocabulary translation of morphologically rich languages." @default.
- W4300692318 created "2022-10-04" @default.
- W4300692318 creator A5015363344 @default.
- W4300692318 creator A5025827839 @default.
- W4300692318 creator A5068340972 @default.
- W4300692318 creator A5086132343 @default.
- W4300692318 date "2017-07-31" @default.
- W4300692318 modified "2023-09-24" @default.
- W4300692318 title "Linguistically Motivated Vocabulary Reduction for Neural Machine Translation from Turkish to English" @default.
- W4300692318 doi "https://doi.org/10.48550/arxiv.1707.09879" @default.
- W4300692318 hasPublicationYear "2017" @default.
- W4300692318 type Work @default.
- W4300692318 citedByCount "0" @default.
- W4300692318 crossrefType "posted-content" @default.
- W4300692318 hasAuthorship W4300692318A5015363344 @default.
- W4300692318 hasAuthorship W4300692318A5025827839 @default.
- W4300692318 hasAuthorship W4300692318A5068340972 @default.
- W4300692318 hasAuthorship W4300692318A5086132343 @default.
- W4300692318 hasBestOaLocation W43006923181 @default.
- W4300692318 hasConcept C111335779 @default.
- W4300692318 hasConcept C138885662 @default.
- W4300692318 hasConcept C149635348 @default.
- W4300692318 hasConcept C154945302 @default.
- W4300692318 hasConcept C186644900 @default.
- W4300692318 hasConcept C203005215 @default.
- W4300692318 hasConcept C204321447 @default.
- W4300692318 hasConcept C2524010 @default.
- W4300692318 hasConcept C2777601683 @default.
- W4300692318 hasConcept C2780513914 @default.
- W4300692318 hasConcept C2781121862 @default.
- W4300692318 hasConcept C28490314 @default.
- W4300692318 hasConcept C33923547 @default.
- W4300692318 hasConcept C41008148 @default.
- W4300692318 hasConcept C41895202 @default.
- W4300692318 hasConcept C80875076 @default.
- W4300692318 hasConcept C90805587 @default.
- W4300692318 hasConceptScore W4300692318C111335779 @default.
- W4300692318 hasConceptScore W4300692318C138885662 @default.
- W4300692318 hasConceptScore W4300692318C149635348 @default.
- W4300692318 hasConceptScore W4300692318C154945302 @default.
- W4300692318 hasConceptScore W4300692318C186644900 @default.
- W4300692318 hasConceptScore W4300692318C203005215 @default.
- W4300692318 hasConceptScore W4300692318C204321447 @default.
- W4300692318 hasConceptScore W4300692318C2524010 @default.
- W4300692318 hasConceptScore W4300692318C2777601683 @default.
- W4300692318 hasConceptScore W4300692318C2780513914 @default.
- W4300692318 hasConceptScore W4300692318C2781121862 @default.
- W4300692318 hasConceptScore W4300692318C28490314 @default.
- W4300692318 hasConceptScore W4300692318C33923547 @default.
- W4300692318 hasConceptScore W4300692318C41008148 @default.
- W4300692318 hasConceptScore W4300692318C41895202 @default.
- W4300692318 hasConceptScore W4300692318C80875076 @default.
- W4300692318 hasConceptScore W4300692318C90805587 @default.
- W4300692318 hasLocation W43006923181 @default.
- W4300692318 hasOpenAccess W4300692318 @default.
- W4300692318 hasPrimaryLocation W43006923181 @default.
- W4300692318 hasRelatedWork W1585034923 @default.
- W4300692318 hasRelatedWork W2083217544 @default.
- W4300692318 hasRelatedWork W2086703572 @default.
- W4300692318 hasRelatedWork W2114358883 @default.
- W4300692318 hasRelatedWork W2403038860 @default.
- W4300692318 hasRelatedWork W2435130738 @default.
- W4300692318 hasRelatedWork W2998280981 @default.
- W4300692318 hasRelatedWork W3107474891 @default.
- W4300692318 hasRelatedWork W363292665 @default.
- W4300692318 hasRelatedWork W40015776 @default.
- W4300692318 isParatext "false" @default.
- W4300692318 isRetracted "false" @default.
- W4300692318 workType "article" @default.