Learning to Translate : A Statistical and Computational Analysis
Joint Authors
De Bie, Tijl
Cristianini, Nello
Goutte, Cyril
Turchi, Marco
Source
Advances in Artificial Intelligence
Issue
Vol. 2012, Issue 2012 (31 Dec. 2012), pp.1-15, 15 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2012-04-22
Country of Publication
Egypt
No. of Pages
15
Main Subjects
Information Technology and Computer Science
Science
Abstract EN
We present an extensive experimental study of Phrase-based Statistical Machine Translation, from the point of view of its learning capabilities.
Very accurate Learning Curves are obtained, using high-performance computing, and extrapolations of the projected performance of the system under different conditions are provided.
Our experiments confirm existing and mostly unpublished beliefs about the learning capabilities of statistical machine translation systems.
We also provide insight into the way statistical machine translation learns from data, including the respective influence of translation and language models, the impact of phrase length on performance, and various unlearning and perturbation analyses.
Our results support and illustrate the fact that performance improves by a constant amount for each doubling of the data, across different language pairs, and different systems.
This fundamental limitation seems to be a direct consequence of Zipf law governing textual data.
Although the rate of improvement may depend on both the data and the estimation method, it is unlikely that the general shape of the learning curve will change without major changes in the modeling and inference phases.
Possible research directions that address this issue include the integration of linguistic rules or the development of active learning procedures.
American Psychological Association (APA)
Turchi, Marco& De Bie, Tijl& Goutte, Cyril& Cristianini, Nello. 2012. Learning to Translate : A Statistical and Computational Analysis. Advances in Artificial Intelligence،Vol. 2012, no. 2012, pp.1-15.
https://search.emarefa.net/detail/BIM-475301
Modern Language Association (MLA)
Turchi, Marco…[et al.]. Learning to Translate : A Statistical and Computational Analysis. Advances in Artificial Intelligence No. 2012 (2012), pp.1-15.
https://search.emarefa.net/detail/BIM-475301
American Medical Association (AMA)
Turchi, Marco& De Bie, Tijl& Goutte, Cyril& Cristianini, Nello. Learning to Translate : A Statistical and Computational Analysis. Advances in Artificial Intelligence. 2012. Vol. 2012, no. 2012, pp.1-15.
https://search.emarefa.net/detail/BIM-475301
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-475301