Improvement of English-Hindi machine translation using ConceptNet

Abstract

The paper is based on the framework of translation of English sentences into Hindi language by using ConceptNet for Statistical Machine Translation and Rule-Based Machine Translation system together. ConceptNet is a knowledge-based graph that represents relation between words using relations like “RelatedTo” and “TranslatedFrom”. Baseline method doesn't appropriately translate the sentences. This work increases the accuracy of translation by using common sense knowledge for translation of those words which are obscured or not translated. We used BLEU score to check the accuracy of our English to Hindi translation system. This approach provides the improved quality of translation.