Pre-processing Steps on Bilingual Corpus for SMT

Main Article Content

Abhijit Paul
Prof. Bipul Syam Purkayastha

Abstract

Correct pre-processing techniques on bilingual parallel corpus (text translated into two different languages) play an important role in the statistical machine translation (SMT) systems. The purpose of this paper is to examine the effectiveness of pre-processing techniques that are used for training the translation systems. This paper also discusses the various pre-processing steps used for SMT. Experimental results of the pre-processing steps on Technology Development for Indian Languages (TDIL) ‘English-Nepali’ parallel corpus (on tourism domain) of size over 4K sentences are discussed.

Downloads

Download data is not yet available.

Article Details

Section
Articles