Abstract

This paper presents a proposed system for machine translation of English interrogative sentences to their Marathi counterpart. The system takes simple interrogative English sentence as an input and performs its lexical analysis. Every token produced by lexical analysis is searched in the English lexicon. If the token is found in the lexicon, its morphological information is retrieved. The local word groups are formed depending on the morphological information of a token, if required. This local word groups are checked against grammar rules of English language for syntactic validity of input sentence. This is done by using bottom up parsing technique. If the input sentence is syntactically correct, the corresponding Marathi token is searched in Marathi Lexicon. If all such tokens corresponding to English tokens are found, then Marathi sentence is produced using Marathi grammar rules. The paper emphasizes on developing production rules for simple English sentences.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call