<![CDATA[Sentiment analysis is a very popular field of research in text mining. The basic idea of ​​sentiment analysis is finding the polarity of the document and classifying it into positive or negative. The text documents used in the research are reviews on the Google Play Store regarding the mobile banking application. Support Vector Machine is a method used and added Lexicon Based Features as additional feature besides using the Bag of Words. The research data is 500 data by dividing 90% training data and 10% test data. The system evaluation results obtained with a combination of Bag of Words and Lexicon Based Features are higher than the results of system evaluations that only use the Bag of Words and systems that only use Lexicon Based Features. The evaluation results obtained by the combination of the two features with testing using 10 fold cross validation are accuracy = 0,846, recall = 0,846, precision = 0,864, and f-measure = 0,855 with the Support Vector Machine parameter value used is the best parameter value of sigma kernel RBF = 3, lambda = 0,1, gamma = 0,001, complexity = 0,1, epsilon = 0,001, and iteration = 50.]]>
Copyrights © 2019
