Part of Speech Tagging for Kayah Language Using Hidden Markov Model
Zar Zar Linn, Pushpa B. Patil · 2019
Part of Speech tagging is one of the active research problem in Natural Language Processing involves language understanding, reasoning and the utilization of common sense of knowledge. This paper contributes the pioneer evaluation to the Kayah Language for tagging Part of Speech. Kayah language is one of the low resource languages in Myanmar. Kayah corpus used in this system is translated by Kayah people from Myanmar. The goal is to build the Kayah Language Part of Speech Tagging System based Hidden Markov Model. Sixteen tag sets are defined for this language. Hidden Markov Model is used to learn the Kayah corpus of words annotated with the correct Part of Speech tags and generated the model relating to the Initial, Transition and Emission probabilities for Kayah Language. Then the generated model is used to decode Viterbi algorithm. This Kayah Hidden Markov Model based POS tagging has been experimented and has overall accuracy of 87 % on test data.