Accession Number : ADA461156


Title :   Statistical Phrase-Based Translation


Corporate Author : UNIVERSITY OF SOUTHERN CALIFORNIA MARINA DEL REY INFORMATION SCIENCES INST


Personal Author(s) : Koehn, Philipp ; Och, Franz J ; Marcu, Daniel


Full Text : http://www.dtic.mil/dtic/tr/fulltext/u2/a461156.pdf


Report Date : Jan 2003


Pagination or Media Count : 8


Abstract : We propose a new phrase-based translation model and decoding algorithm that enables us to evaluate and compare several, previously proposed phrase-based translation models. Within our framework, we carry out a large number of experiments to understand better and explain why phase-based models out-performed word-based models. Our empirical results, which hold for all examined language pairs, suggest that the highest levels of performance can be obtained through relatively simple means: heuristic learning of phrase translations from word-based alignments and lexical weighting of phrase translations. Surprisingly, learning phrases longer than three words and learning phrases from high-accuracy word-level alignment models does not have a strong impact on performance. Learning only syntactically motivated phrases degrades the performance of our systems.


Descriptors :   *ALGORITHMS , *TRANSLATIONS , *PHRASE STRUCTURE GRAMMARS , *DECODING , HEURISTIC METHODS , LEXICOGRAPHY , LEARNING , WORDS(LANGUAGE)


Subject Categories : Linguistics
      Numerical Mathematics


Distribution Statement : APPROVED FOR PUBLIC RELEASE