Kamba Part of Speech Tagger Using Memory Based Approach
Benson Kituku, Musumba George, Peter Waiganjo Wagacha · International Journal on Natural Language Computing · 2015
Part of speech tagging is very important and the initial work towards machine translation and text manipulation.Though much has been done in this regard to the Indo-European and Asiatic languages, development of part of speech tagging tools for African languages is wanting.As a result, these languages are classified as under resourced languages.This paper presents data driven part of speech tagging tools for kikamba which is an under resourced language spoken mostly in Machakos, Makueni and Kitui.The tool is made using the lazy learner called Memory Based Tagger (MBT) with approximately thirty thousand word corpuses.The corpus is collected, cleaned and formatted with regard to MBT and experiment run.Very encouraging performance is reported despite little amount of corpus, which clearly shows that using the state of art technology of data driven methods tools can be developed for under resourced languages.We report a precision of 83%, recall of 72% and F-score of 75% and in terms of accuracy for the known and unknown words, and accuracy of 94.65% and71.93%respectively with overall accuracy of 90.68%..This predicts that with little source of corpus using data driven approach, we can generate tools for the under resourced languages in Kenya.