Example based English to Bengali machine translation

Abstract

In this thesis we propose a new architecture for example based English to Bengali machine translation. The proposed Example Based Machine Translation (EBMT) system has five steps: 1) Tagging 2) Parsing 3) Prepare the chunks of the sentence using sub-sentential EBMT 4) Using an efficient adapting scheme match the sentence rule 5) Translate from English to Bengali in the chunk and generate output with morphological analysis. We prepared our tag set for tagging the English sentence. Here we proposed an optimal adapting scheme for choosing sentence rule from the knowledge base of the EBMT system. Our current system can translate simple sentences. We also defined a way to translate a complex sentence using sub-sentential EBMT. As this system can add more rules in the knowledge base, eventually it can be use for general purpose English to Bengali machine translation.