Comparison and system combination of n-gram-based and syntax-based machine translation systems

In this paper we shall compare two approaches to machine translation: the Syntax Augmented Machine Translation system (SAMT), which is a syntaxdriven translation system, underlain by phrase-based model, and the n-gram-based Statistical Machine Translation (SMT), in which a translation process is based on statistical modeling of the bilingual context. We provide a step-by-step comparison of the systems, reporting results in terms of automatic evaluation metrics and required computational resources for a smaller Arabic-to-English translation task from the news domain. Finally, we combine the output of both systems that yield to significant improvement of translation quality.