Comparing SVM and KNN Algorithms for Myanmar News Sentiment Analysis System
Thein Yu, Khin Thandar Nwet · 2020
Sentiment analysis is one of the natural language processing research fields that identify the polarity and subjectivity of documents. With the increasing of web technology, large volumes of data are available from many resources. Sentiment analysis applications are widely applied in many domains from news, financial, election, blogs, and post. News is very important and provides valuable information for society. News is tagged as positive and negative in this system. News is collected from many websites and ALT tree bank. N-gram and TF-IDF feature extraction and selection methods are used in this system to get more performance. Supervised machine learning algorithm is a classification algorithm that uses labeled data. K nearest neighbors is a simple algorithm that stores all available features and classifies new features based on a similarity measure. SVM is one of the classifier that has higher accuracy. This paper shows comparison of performance results for sentiment analysis system by using support vector machine (SVM) and K nearest neighbor (KNN) algorithms.