This study conducted topic modeling on user reviews of the Sapawarga – Jabar Super App application on the Google Play Store. The dataset comprised user reviews submitted since the date application released until January 7, 2026 yielded 5.082 user reviews. Collected data processed through a series of data processing pipeline namely case folding, remove punctuation, normalization, tokenization, stop word removal and stemming. Topic modeling was conducted using Latent Dirichlet Allocation (LDA) algorithm grouped based on year of data. Topic distribution was visualized using the PyLDAvis to facilitate analysis and interpretation. This study found there are 9 topic after analysing year by year data namely Appreciation of the usefulness of the application in helping residents (1), Technical issues in terms of how to log in to the application (2), Application improvement proposal (3), Ease of information and citizen empowerment (4), General expressions of appreciation and criticism from citizens towards the application (5), Technical issues with payment features (6), Application error problem (7), User appreciation for the ease of vehicle tax payments from the application (8) and Criticism of application constraints in vehicle tax payments (9). The interpretation results indicate that the application is widely used by residents for online tax payments, particularly vehicle tax services, compared to other features or uses of the application. The findings of this study are expected to serve as a valuable insight for relevant stakeholders to improve the quality, performance, and overall effectiveness of the application as a public service platform.
Copyrights © 2026