Sentiment Analysis System for Myanmar News using K Nearest Neighbor and Naïve Bayes
Abstract— With the explosive growth of internet technology, there are very large amount of information on the web for the internet users. Users not only use that information but also provide opinions for decision making process. Sentiment analysis or opinion mining is one of text classification techniques that identify and extract opinion described in a piece of text. Our aims in this paper are to develop automatic sentiment analysis system for Myanmar news and to annotate sentiment news. Therefore, this system creates sentiment annotated corpus for Myanmar news. Feature extraction and selection are very important for sentiment analysis to get higher performance. N-grams, Countvectorizer, and TF-IDF are used for feature selection and feature extraction. In this system, Myanmar news sentiment analysis system is implemented by using K Nearest Neighbor (KNN) and Naïve Bayes machine learning algorithms.
Index Terms— Sentiment analysis, Naïve Bayes, K Nearest Neighbor, N-gram, TF-IDF
University of Computer Studies, MYANMAR
Khin Thandar Nwet
University of Information Technology, MYANMAR
Cite: Thein Yu, Khin Thandar Nwet, "Sentiment Analysis System for Myanmar News using K Nearest Neighbor and Naïve Bayest," Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering (WCSE 2020), pp. 1-5, Yangon (Rangoon), Myanmar (Burma), February 26-28, 2020.