Translation Divergence Patterns Handling in English to Urdu Machine Translation

Abstract

English to Urdu machine translation is still in its infancy. This study illustrates various types of translation divergences and their implication in English to Urdu machine translation. The divergence in English to Urdu machine translation can be thought as representing the translation divergences between Subject-Verb-Object (SVO) class languages to Subject-Object-Verb (SOV) class languages. This study discusses the different types of divergences in English to Urdu machine translation and presents novel computational algorithms to detect and to resolve these divergences in English to Urdu Machine Translation. These algorithms for detection of divergence have been implemented in English to Urdu Machine Translation system, and results have been presented in this paper. The work introduced here is the only one, to the best of our knowledge, which automatically detects and resolves divergences in English to Urdu machine translation.