FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2768040551> ?p ?o ?g. }

• W2768040551 abstract "Statistical Machine Translation has been shown to benefit from complex linguistic structures. However, previous work mainly focuses on sequences and trees. In this thesis, we build dependency graphs which are constructed from dependency trees and uniformly represent both dependency relations and sequential relations, including bigram relations and sibling relations. We propose translation models to translate these graphs into target strings and conduct experiments on Chinese--English and German--English translation tasks.As a motivation, we firstly present a pseudo forest-to-string model which improves a dependency tree-to-string model by dependency decomposition. The decomposition takes sibling relations into consideration which results in more rules being used and thus a higher phrase coverage. Experiments show that such decomposition is beneficial to translation performance. Integrating phrasal rules further improves our model.Then, we propose a segmentational graph-based translation model. It segments graphs into subgraphs and generates translations from left to right by combining translations of these subgraphs. The graphs explicitly combine dependency relations and bigram relations. In experiments, the graph-based model outperforms both the phrase-based model and treelet-based model. In addition, we improve this model by using a graph segmentation model to take source context into consideration.Furthermore, inspired by using tree grammars to translate trees, we propose recursive graph-based translation models by using graph grammars. An edge replacement grammar is used to translate dependency-edge graphs which are converted from dependency trees by labeling edges to naturally take sibling relations into consideration. A node replacement grammar is used to translate dependency-sibling graphs which explicitly add sibling links to dependency trees. Experiments show that our models are significantly better than the hierarchical phrase-based model." @default.
• W2768040551 created "2017-11-17" @default.
• W2768040551 creator A5038537171 @default.
• W2768040551 date "2016-11-01" @default.
• W2768040551 modified "2023-09-26" @default.
• W2768040551 title "Dependency graph-based statistical machine translation" @default.
• W2768040551 cites W1502293651 @default.
• W2768040551 cites W1508113248 @default.
• W2768040551 cites W1512742637 @default.
• W2768040551 cites W1517947178 @default.
• W2768040551 cites W1551202288 @default.
• W2768040551 cites W1555286493 @default.
• W2768040551 cites W1585340632 @default.
• W2768040551 cites W1631260214 @default.
• W2768040551 cites W171828413 @default.
• W2768040551 cites W1829822087 @default.
• W2768040551 cites W1916559533 @default.
• W2768040551 cites W1964388487 @default.
• W2768040551 cites W1970961429 @default.
• W2768040551 cites W2001064229 @default.
• W2768040551 cites W2006969979 @default.
• W2768040551 cites W2048390999 @default.
• W2768040551 cites W2048808408 @default.
• W2768040551 cites W2049633694 @default.
• W2768040551 cites W2060127787 @default.
• W2768040551 cites W2073955506 @default.
• W2768040551 cites W2088602194 @default.
• W2768040551 cites W2095690342 @default.
• W2768040551 cites W2095755718 @default.
• W2768040551 cites W2096175520 @default.
• W2768040551 cites W2097333193 @default.
• W2768040551 cites W2101105183 @default.
• W2768040551 cites W2102258849 @default.
• W2768040551 cites W2106310992 @default.
• W2768040551 cites W2107038437 @default.
• W2768040551 cites W2108540317 @default.
• W2768040551 cites W2110104386 @default.
• W2768040551 cites W2112900913 @default.
• W2768040551 cites W2113541941 @default.
• W2768040551 cites W2116316001 @default.
• W2768040551 cites W2116492146 @default.
• W2768040551 cites W2118536060 @default.
• W2768040551 cites W2119168550 @default.
• W2768040551 cites W2124807415 @default.
• W2768040551 cites W2125474347 @default.
• W2768040551 cites W2127191681 @default.
• W2768040551 cites W2131988669 @default.
• W2768040551 cites W2132382844 @default.
• W2768040551 cites W2132622673 @default.
• W2768040551 cites W2133956246 @default.
• W2768040551 cites W2137076533 @default.
• W2768040551 cites W2138395083 @default.
• W2768040551 cites W2144600658 @default.
• W2768040551 cites W2145752261 @default.
• W2768040551 cites W2149327368 @default.
• W2768040551 cites W2149837184 @default.
• W2768040551 cites W2152263452 @default.
• W2768040551 cites W2152752344 @default.
• W2768040551 cites W2153653739 @default.
• W2768040551 cites W2153800732 @default.
• W2768040551 cites W2154124206 @default.
• W2768040551 cites W2154581043 @default.
• W2768040551 cites W2156985047 @default.
• W2768040551 cites W2158195707 @default.
• W2768040551 cites W2159755860 @default.
• W2768040551 cites W2160382364 @default.
• W2768040551 cites W2161795601 @default.
• W2768040551 cites W2164415172 @default.
• W2768040551 cites W2165666205 @default.
• W2768040551 cites W2166451556 @default.
• W2768040551 cites W2168966090 @default.
• W2768040551 cites W2169237085 @default.
• W2768040551 cites W2250623140 @default.
• W2768040551 cites W2250907725 @default.
• W2768040551 cites W2251053286 @default.
• W2768040551 cites W2252123671 @default.
• W2768040551 cites W2296308987 @default.
• W2768040551 cites W230880734 @default.
• W2768040551 cites W2401082558 @default.
• W2768040551 cites W2437005631 @default.
• W2768040551 cites W2964308564 @default.
• W2768040551 cites W31462306 @default.
• W2768040551 cites W75781568 @default.
• W2768040551 cites W76590478 @default.
• W2768040551 cites W2109776855 @default.
• W2768040551 cites W2114396466 @default.
• W2768040551 hasPublicationYear "2016" @default.
• W2768040551 type Work @default.
• W2768040551 sameAs 2768040551 @default.
• W2768040551 citedByCount "1" @default.
• W2768040551 countsByYear W27680405512021 @default.
• W2768040551 crossrefType "dissertation" @default.
• W2768040551 hasAuthorship W2768040551A5038537171 @default.
• W2768040551 hasConcept C108757681 @default.
• W2768040551 hasConcept C132525143 @default.
• W2768040551 hasConcept C137546455 @default.
• W2768040551 hasConcept C154945302 @default.
• W2768040551 hasConcept C157486923 @default.
• W2768040551 hasConcept C16311509 @default.
• W2768040551 hasConcept C19768560 @default.

• next