Analysis and System Combination of Phrase- and N-Gram-Based Statistical Machine Translation Systems

In the framework of the Tc-Star project, we analyze and propose a combination of two Statistical Machine Translation systems: a phrase-based and an N-gram-based one. The exhaustive analysis includes a comparison of the translation models in terms of efficiency (number of translation units used in the search and computational time) and an examination of the errors in each system’s output. Additionally, we combine both systems, showing accuracy improvements.

• Type: Conference
• Year: 2007
• Where: NAACL
• Authors: Marta R. Costa-Jussà, Josep Maria Crego, David Vilar, José A. R. Fonollosa, José B. Mariño, Hermann Ney
• Keywords: Computational Linguistics, NAACL 2007, Statistical Machine Translation, Tc-Star Project, Translation Models