Building a parallel Corpora: Translation issues and remedial case

In a multilingual country like India, language translation plays a prominent role in the field of text processing and applications, such as, machine learning, machine translation, information extraction, information retrieval, and natural language understanding systems. There are many issues and challenges in translation between languages such as lexical divergences, ambiguities, lexical mismatches, reordering, syntactic and semantic issues, structural changes etc. In this paper, we have presented a comparative analysis by highlighting some of the crucial issues, challenges and solutions in translation from Hindi to Malayalam language, while executing Indian Language Corpora Initiative (ILCI) project of DeitY, GoI. In this analysis, we have found that Malayalam and Hindi have very distinct structures in linguistic levels like phonological, lexical and syntactic, which has to be handled carefully in translation.