Thesis (Selection of subject)Thesis (Selection of subject)(version: 385)
Thesis details
   Login via CAS
Combining outputs from machine translation systems
Thesis title in Czech:
Thesis title in English: Combining outputs from machine translation systems
English key words: Machine Translation, Combination, Confusion Network Decoding, Sentence re-ranking
Academic year of topic announcement: 2009/2010
Thesis type: diploma thesis
Thesis language: angličtina
Department: Institute of Formal and Applied Linguistics (32-UFAL)
Supervisor: prof. Ing. Zdeněk Žabokrtský, Ph.D.
Author: hidden - assigned and confirmed by the Study Dept.
Date of registration: 10.11.2009
Date of assignment: 11.11.2009
Date and time of defence: 04.02.2011 00:00
Date of electronic submission:16.12.2010
Date of submission of printed version:20.12.2010
Date of proceeded defence: 04.02.2011
Opponents: doc. RNDr. Ondřej Bojar, Ph.D.
 
 
 
Guidelines
Combination of multiple machine translation (MT) systems
has emerged as a powerful method of combining their strength,
leading to better translation quality in comparison
to the systems in isolation.

The goal of this thesis is to design and implement a software
tool capable of combining outputs of at least three different MT
systems. The selected combination approach should be as language
and system independent as possible. Translation quality
gain will be evaluated on the task of English-to-Czech MT using
state-of-the-art translation systems and automatic translation
metrics.

References
[1] Xiaodong He and Kristina Toutanova: Joint Optimization for Machine
Translation System Combination. In Proceedings of EMNLP. 2009

[2] Antti-Veikko I. Rosti, Bing Xiang, Spyros Matsoukas, Richard Schwartz,
Necip Fazil Ayan, and Bonnie J. Dorr: Combining outputs from
multiple machine translation systems. In Proc. of NAACL-HLT,
pp. 228--235. 2007.

[3] Fei Huang, Kishore Papineni: Hierarchical System Combination for
Machine Translation. Proceedings of the 2007 Joint Conference on
EMNLP, pp. 277-286, Prague, 2007

[4] Josh Schroeder, Trevor Cohn, and Philipp Koehn: Word Lattices for
Multi-Source Translation. Proceedings of the 12th Conference of the
European Chapter of the ACL, pages 719-727, Athens, Greece, 2009.
 
Charles University | Information system of Charles University | http://www.cuni.cz/UKEN-329.html