We developed English-Bangla parallel corpora for statistical machine
translation. By hand we tagged 20,000 words of our Bangla corpus according to their
particular part of speeches. In our work we also suggested a method for identifying
word correspondence in parallel English-Bangla text using a translation model based
on part of speech and n-gram model.
This thesis report is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2008.
Cataloged from PDF version of thesis report.
Includes bibliographical references (page 93).
Department:
Department of Computer Science and Engineering, BRAC University
