Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Journal of Information Systems and Informatics
Dias Henandra Eka Putra
Muria Kudus University
Computer Science & IT

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

Found 1 Documents
Search

 1 
Time-Series Monitoring of Sentiment Dynamics in Reviews of Four Indonesian E-Wallet Applications Using a Hybrid TF-IDF and Bi-LSTM Framework Noor Latifah; Dias Henandra Eka Putra; Fajar Nugraha
Journal of Information System and Informatics Vol 8 No 2 (2026): April
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i2.1488

Abstract

This study proposes a hybrid sentiment analysis framework to examine user perceptions of four Indonesian e-wallet applications using Google Play Store reviews. The framework combines TF-IDF features reduced through Truncated SVD with a Bidirectional Long Short-Term Memory (Bi-LSTM) model within a two-stage evaluation design consisting of holdout classification and external temporal inference. For supervised classification, 20,000 raw reviews were filtered and labeled using a rating-based strategy, resulting in 13,823 labeled reviews. Reviews with ratings of 4–5 stars were assigned to the positive class and 1–2 stars to the negative class; these labels should be interpreted as sentiment proxies rather than fully human-validated ground truth. A second dataset of 24,000 reviews was constructed for balanced cross-application temporal comparison across 2024–2026. On the holdout test set, the proposed model achieved an accuracy of 0.881, with macro-F1 and weighted-F1 scores of 0.881. Under the external temporal setting, DANA remained relatively stable, GoPay improved markedly in 2025 and remained high in 2026, ShopeePay showed a gradual decline, and OVO exhibited the strongest negative trend. These results indicate that the proposed framework is useful not only for supervised sentiment classification but also for structured temporal monitoring across e-wallet platforms.
Co-Authors Fajar Nugraha Noor Latifah