<![CDATA[Mobile JKN application is a form of BPJS Kesehatan's commitment in providing services and ease of access for BPJS Kesehatan users. BPJS Kesehatan in organizing the health insurance program since 2014, can be assessed how the people of Indonesia make use of health insurance implementation facilities through the JKN Mobile application based on user reviews of the application. Sentiment analysis needs to be done to analyze reviews provided by app user. This study used the Maximum Entropy classification method coupled with the Gini Index Text for feature selection. Sentiment analysis consists of data collection process, text preprocessing, word weighting with raw tf, followed by feature selection using Gini Index Text, then classification using Maximum Entropy with features obtained from the previous feature selection. The results of this study are that the best accuracy value is obtained when using the number of features or threshold of 80%, with a value of evaluation as an accuracy of 85,36%, a precision of 92,18%, a recall of 75,59%, and f-measure of 82,85%.]]>
Copyrights © 2021
