期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2011
卷号：2
期号：4
页码：1441-1447
出版社：TechScience Publications
摘要：This paper deals about the transliteration of the identified Multiword Expression (MWE) of Manipuri using Conditional Random Field (CRF). Manipuri is a very highly agglutinative language and is an Eight Scheduled Language of Indian Constitution. This language uses multiple script (two scripts); the first one is purely of its own origin called Meitei Mayek(Script) while another one is a borrowed Bengali Script. The very nature of resource constraint for the Meitei Script comparing to Bengali Script Manipuri compels us to think of transliteration to the output of MWE identification as another means for MWE identification in Meitei Script Manipuri. MWE plays an important role in the applications of Natural Language Processing like Machine Translation, Part of Speech tagging, Information Retrieval, Question Answering etc. Feature selection is an important factor in recognition of Manipuri MWE using CRF. This model proved to have the Recall (R) of 64.08%, Precision (P) of 86.84% and F-measure (F) of 73.74%. The transliterated output has an accuracy of 90.01% when compare with both the output of Meitei Script to Bengali Script Manipuri.b
