Tribune News Service
Patiala, October 30
A research scholar at Punjabi University has prepared an English and Punjabi parallel comparable corpus as part of his research work. University officials said this was the first such language resource for English and Punjabi, which would be beneficial in research practice in relation to translation work.
Manpreet Singh is carrying out research under the guidance of Dr Ajit Kumar, associate professor, Department of Computer Science, Multani Mal Modi College, and Dr Vishal Goyal, professor, Department of Computer Science, Punjabi University. He is working as an assistant professor at a Jalandhar college.
“The newly developed system focuses on extracting parallel data and building up a corpus. It takes two input files of English and Punjabi each and pairs similar sentences of both files as output. This will compensate for the lack of parallel data, which can be used for statistical machine translation tasks in English and Punjabi,” said Manpreet.
He has collected 15 lakh sentences of the two languages from Wikipedia and official websites on language resources, which would help researchers working in the field, added Manpreet.
“The software developed can be modified to be used for other language pairs as well,” said Manpreet. Dr Vishal Goyal said the tool would be useful for those, who want to do research in relation to translation from English to Punjabi and vice-versa. “Parallel corpora are the backbone of deep learning and machine learning,” he added.
Wants elaborate cold storage facilities for inoculation | Te...
CM says may go in for stricter curbs | Makeshift hospitals t...
CM warns of strict action if protesters turn violent