Date of Publication :27th December 2022
Abstract: The Part of Speech (POS) tagging tries to tag each word with its correct part of speech. In this paper we discuss about the rule-based POS tagger for Kokborok, a resource poor and less digitized Indian language. We employ two machine learning algorithms for supervised approach, Naive Bayes (NB) and decision tree. Experimental results showed rule-based approach performed better than Naive Bayes and Decision tree from supervised approach giving accuracy of 79%,70% and 71% respectively.
Reference :
Will Updated soon