Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Journal of Renewable Energy, Electrical, and Computer Engineering
Putri, Nabila Rizky Amalia
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Energy Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Sentiment Analysis on Digital Korlantas POLRI Application Reviews Using the Distilbert Model Putri, Nabila Rizky Amalia; Trimono, Trimono; Damaliana, Aviolla Terza
Journal of Renewable Energy, Electrical, and Computer Engineering Vol. 4 No. 2 (2024): September 2024
Publisher : Institute for Research and Community Service (LPPM), Universitas Malikussaleh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/jreece.v4i2.17197

Abstract

The implementation of digitalization in public services by Korlantas Polri has facilitated faster administration, wider access, and improved service quality. The Korlantas Polri Digital app has garnered more than 5 million downloads on the Google Play Store, with a rating of 3.7 and around 110 thousand reviews. Given that an app's reputation can be significantly affected by criticism, sentiment analysis becomes very important to categorize user reviews as positive, negative, or neutral, thus assisting developers in identifying app shortcomings. This study uses DistilBERT, a deep learning model distilled from BERT, to assess the effectiveness of sentiment analysis on reviews. Data was collected from user reviews on the Google Play Store between September 1, 2023 and May 31, 2024, resulting in 8,752 reviews retained for analysis. Model performance was evaluated at three data ratios: 60:40, 70:30, and 80:20, with the best performance results seen at a ratio of 80:20, achieving 88% accuracy. Increasing the training data ratio from 60:20 to 80:20 has a positive impact on the model, suggesting that the model can learn better with larger training data.
Co-Authors Damaliana, Aviolla Terza Trimono, Trimono