ANALYSIS OF OIL SENTIMENT SENTIMENTS ON TWITTER USING SUPPORT VECTOR MACHINE

Twitter is one of the social media platforms used by people in Indonesia. often its users to express opinions regarding a product, institution or event. From keyword fuel, fuel subsidy that currently trending topic because changes subsidies affect prices other staples, find out value sentiment public opinion, analysis methods support vector machine and lexicon based. 
 Lexicon labeling method matching words contained document with dictionary. After labeling, data tested using classification method, stage carried after going through preprocessing phase, where tweet results tend be positive negative, Support Vector Machine validated K-Fold Cross Validation.This research produced 50,001 which were divided into 21,561 sentiments, 9206 neutral sentiments 19234 negative sentiments. these it can concluded shows for rising changing prices.