Paper Title
N-Gram Machine Translation System For English To Punjabi Transliteration
Abstract
In this paper the system is made in such a way that it first learn from the existing examples stored in the database
and then with the help of N-Gram approach we translate the proper nouns of English language into its equivalent
Gurumukhi Script. In N-Gram approach we have used unigram, bi-gram, tri-gram, four-gram, five-gram and six – gram for
transliteration purpose. We store the unigrams in database manually and system is proposed to extract bi-gram, tri-gram,
four-gram, five-gram and six – gram and then store it into the database along with their frequency of occurrence we are
presenting Statistical Machine Translation system to Transliterate Proper nouns written in English language into its
equivalent Punjabi language. The main purpose of transliteration is removing the language barrier for communication. It can
also use various information retrieval systems. Transliteration is a process to generate the words from the source language to
the target language while keeping the pronunciation of the source and target words equivalent. The reverse process is know
as backward transliteration. This system can be used in various government organizations in India because English is used as
communication language in almost all government office. The proposed system is divided to two parts – learning and
transliteration.
Keywords- Transliteration, Statistical Machine Translation, English language, Gurumukhi Script , Machine Translation.