Year: 2025
Type: article
Abstract: E-money usage in Indonesia has grown significantly due to increasing internet penetration and smartphone adoption. Digital transactions are becoming more common, with platforms like GoPay, OVO, and Dana leading the market. The government and financial institutions actively support this shift through regulations and initiatives. This study analyzes user sentiment on the Pospay application using the BERT deep learning method, based on 16,760 Google Play Store reviews. To the best of our knowledge, this is the first study to apply BERT for sentiment analysis of Pospay user reviews in Indonesia. The goal is to understand user perceptions and satisfaction. BERT helps capture subtle nuances in reviews, including informal expressions and abbreviations like "gk" for negative sentiment. The model achieves high accuracy, with precision scores of 0.82 (negative) and 0.93 (positive), and recall scores of 0.92 (negative) and 0.93 (positive). Findings suggest PT Pos should enhance application stability, security, transaction processing, and customer service. Regular updates are recommended to improve performance and user satisfaction.
Source: JURNAL TEKNIK INFORMATIKA
Authors: Erna Mulyati, Muhammad Ibnu Choldun Rachmatullah, Adri Sapta Firmansyah
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4415095993
FWCI:
Cites:
Cited by:
Related to: