<![CDATA[Digital transformation is driven by the increasing number of internet and mobile phone users in Indonesia, including public services such as the PLN Mobile application. The purpose of this study is to evaluate user sentiment towards PLN Mobile application reviews and find gaps between user ratings and reviews. Through web scraping on Google Play Store, with a total review data of 11,004 reviews between January 2022 and December 2023. During the preprocessing step, SentiStrength_id was used as the labeling approach, and Support Vector Machine was used for modeling. A spelling corrector using Peter Norvig was applied to correct spelling issues. The accuracy of sentiment analysis was much better with this procedure, reaching 82% at a data split ratio of 90:10. The percentage of sentiment obtained was 16.5% negative, 16.1% neutral, and 67.4% positive. The percentage of mismatched user ratings and reviews was 23.1% for negative reviews, 4.5% for neutral, and 72.49% for positive reviews.]]>
Copyrights © 2025
