标题：SENTIMENT ANALYSIS REVIEW ON TWITTER ABOUT THE SUPREME COURT REPUBLIC OF INDONESIA USING SUPPORT VECTOR MACHINE ALGORITHM BASED PARTICLE SWARM OPTIMIZATION AS FEATURE SELECTION METHODS
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：20
出版社：Journal of Theoretical and Applied
摘要：The Supreme Court of the Republic of Indonesia has sought to develop a positive image of the judiciary through various programs. Sentiment analysis constituted a popular research area even in sentiment analysis of text content on twitter. The Support Vector Machine (SVM) classification algorithm was proposed by many researchers to be used in the review sentiment analysis. SVM is able to identify the separated hyperplane that maximizes the margin between two different classes. However, SVM has a weakness for parameter selection or suitable feature. Feature selection can be used to reduce the less relevant attributes on the dataset. The feature selection algorithm used is Particle Swarm Optimization (PSO) to solve the optimization problem in order to increase the classifications accuracy Support Vector Machine. This research generate text classification in the positive or negative of The Supreme Court of the Republic of Indonesia review on twitter. The evaluation was done by using 10-Fold Cross Validation and the measurement accuracy is measured by Confusion Matrix and ROC curve. The result showed an increasing in accuracy SVM of 91.05% to 91.72%.
关键词：Sentiment Analysis; Twitter; SVM; PSO; Feature Selection; Classification Text
