Institute of Formal and Applied Linguistics

The talk will focus on my PhD work titled "English-Bhojpuri SMT System-Insights from the Karaka Model" which was received in March this year.  The talk will focus on the PhD's following points:
(a) improving the accuracy and fluency of the low-resource based Machine Translation system (especially based on statistical method) for Indian language (b) language technology resources such as Universal Dependency and Paninian Dependency based annotated monolingual and parallel corpus for Bhojpuri (c) encoding of Karaka model into Statistical Machine Translation model, and (d) suitability of Paninian Dependency and Universal Dependency for English-Indian languages using Statistical Machine Translation.