Paper Info
Title
Learning Rules to Improve a Machine Translation System
Abstract
In this paper we show how to learn rules to improve the performance of a machine translation system. Given a system consisting of two translation functions (one from language A to language B and one from B to A), training text is translated from A to B and back again to A. Using these two translations, differences in knowledge between the two translation functions are identified, and rules are learned to improve the functions. Context-independent rules are learned where the information suggests only a single possible translation for a word. When there are multiple alternate translations for a word, a likelihood ratio test is used to identify words that co-occur with each case significantly. These words are then used as context in context-dependent rules. Applied on the Pan American Health Organization corpus of 20,084 sentences, the learned rules improve the understandability of the translation produced by the SDL International engine on 78% of sentences, with high precision.
Year
DOI
Venue
2003
10.1007/978-3-540-39857-8_20
Lecture Notes in Artificial Intelligence
Keywords
Field
DocType
context dependent,likelihood ratio test,machine translation
Rule-based machine translation,Example-based machine translation,Computer science,Machine translation,Speech recognition,Synchronous context-free grammar,Machine translation software usability,Transfer-based machine translation,Natural language processing,Artificial intelligence,Computer-assisted translation,Sentence
Conference
Volume
ISSN
Citations 
2837
0302-9743
2
PageRank 
References 
Authors
0.53
11
2
Name
Order
Citations
PageRank
David Kauchak136325.92
Charles Elkan25118572.94