Myanmar Part Of Speech Tagging using Backpropagation Neural Network

Abstract

POS tagging is the process of automatic assigning a for each word with their categories that best suits the definition of the word as well as the context of the sentence of a natural language in which it is used. It is also a fundamental stage in most natural language processing (NLP) tasks. There are different approaches to the problem of POS tagging. This paper describes the usage of BPNN for Myanmar POS tagging. Experiments show that our analysis achieves a good result with simple sentences and complex sentences. Experimental results are showed that on input data representation by 3-gram, 4-gram and 5- gram format. Open and close data are used to evaluate performance. 3-gram format for close data and 4-gram format for open data are more suitable and good results for POS tagging.