Sentiment analysis of economic news in Bahasa Indonesia using majority vote classifier

Internet acts as a media to publicite news, which divided into various categories, such as politics, economics, sports, and health. Each has tendency to publish the news with positive or negative sentiment. News sentiment has ability to affect people's opinion about a topic or government policy. Economy is one interesting topic which has direct impact to Indonesian citizen. Hence, this research focused on economic news from Indonesian online media. This research tested the performance of some classifiers to classify the news in Bahasa Indonesia from various online news media. Popular classifiers such as decision tree, random forests, and support vector machine will be compared each other. After that, they will be combined using majority vote classifier. Majority vote classifier votes the result between them, the voted result then compared again with the other three classifiers. Experiments has been made by this research by combining parameters and classifiers, concluded that combining multiple classifier using Majority Vote scoring better precision and accuracy than decision tree, random forests, and support vector machine alone.