<![CDATA[Apparently Social media sites are becoming increasingly popular, it creates platforms through which organizations, communities, and individuals share and discuss various topics. The reviews and data obtained from these sites are essential for further analysis. In this paper we studied the sentiment classification of 2019 Kenyan 1000 banknote demonetization using Twitter as our source dataset. We perform Multi nominal naïve Bayes classifier algorithm to classify tweets documents. We split our dataset using k-folder validation since we had limited amounts of data, so to achieve unbiased prediction of the model. . We obtained in test data an accuracy of 70.8% when we used unigram model and 64.1% when we applied bigram model. Results show that the model reached to an acceptable accuracy of (71%) on average using unigram model. ]]>
Copyrights © 2019
