Divergence patterns between English and Sanskrit Machine Translation

Abstract

The translation divergence is a challenging problem in the area of machine  translation. A detailed study of divergence issues in machine translation is required for their proper classification and detection. The language divergences between English and Sanskrit can be considered as representing the divergences between SVO (Subject - Verb - Object) and SOV (Subject - Object - Verb) classes of languages. This topic needs exploration to identify different sources of translation divergence between English and Sanskrit. This paper discusses translation patterns between English and Sanskrit to identifying the potential topics of translation divergences. The typical type (specific to language pair such as English and Sanskrit) of divergence is based on different aspects such as linguistic to socio-and psycho-linguistic, role of conjunctions and particles, participle, gerunds and socio-cultural aspects. We have proposed the detection rules for these types of divergence related sentences and apply the adaptation rules on it. The results of divergence are shown in GUI form. We evaluate the results of our system with different evaluation methods of machine translation.