<![CDATA[The public's dependence on fuel oil (BBM) makes it a vital component in supporting the smooth production and distribution of goods and services, as well as maintaining economic stability. However, it is estimated that around 30 percent of energy subsidies have been enjoyed by groups that are not classified as poor or vulnerable. To address this issue, PT Pertamina (Persero) has implemented a new policy requiring vehicle registration through the MyPertamina app for users of subsidized fuels such as Pertalite and Solar. However, many users have expressed complaints regarding the operational aspects of the app. Therefore, this study employs a descriptive quantitative method relying on objective measurements and mathematical analysis to identify user complaints through sentiment analysis using the Long Short-Term Memory (LSTM) machine learning method. The data used consists of 2,000 user reviews of the MyPertamina app obtained from the Google Play Store. The analysis process was conducted using the Python programming language via the Google Colab platform. The results of the study indicate that the majority of user sentiment is negative, suggesting that the quality of MyPertamina's services is still suboptimal. Additionally, this study adopts the five dimensions of e-ServQual to identify factors influencing users' perceptions and sentiment toward the app. These findings are expected to serve as evaluation material for improving MyPertamina's operational services in the future.]]>
Copyrights © 2025
