Pahlevi, Muhammad Firza (2026) Analisis Sentimen Multi-Aspek pada Ulasan Aplikasi MyBCA di Google Playstore Menggunakan IndoBERT dan Kerangka PACMAD. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
22082010094.-cover.pdf Download (4MB) | Preview |
|
|
Text (BAB I)
22082010094.-bab1.pdf Download (319kB) | Preview |
|
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (BAB II)
22082010094.-bab2.pdf Restricted to Repository staff only until 24 May 2028. Download (304kB) |
|
|
Text (BAB III)
22082010094.-bab3.pdf Restricted to Repository staff only until 24 May 2028. Download (400kB) |
|
|
Text (BAB IV)
22082010094.-bab4.pdf Restricted to Repository staff only until 24 May 2028. Download (1MB) |
|
|
Text (BAB V)
22082010094.-bab5.pdf Download (234kB) | Preview |
|
|
Text (DAFTAR PUSTAKA)
22082010094.-daftarpustaka.pdf Download (187kB) | Preview |
|
|
Text (LAMPIRAN)
22082010094.-lampiran.pdf Restricted to Repository staff only until 24 May 2028. Download (221kB) |
Abstract
The rapid growth of digital transformation has increased the demand for fast and convenient digital banking transactions. However, the myBCA application still faces challenges in user adoption and system reliability issues that may affect user trust. This study applies Aspect-Based Sentiment Analysis to comprehensively evaluate the service quality of the myBCA application using an adapted People At the Centre of Mobile Application Development (PACMAD) framework. User reviews were collected from Google Play Store for myBCA versions 2.6.0 to 2.9.0 covering the aspects of Effectiveness, Efficiency, Satisfaction, Learnability, Errors, and Cognitive Load. The modeling process was carried out using IndoBERT through a series of Hyperparameter Tuning and Imbalanced Data handling experiments. The results show that the best model configuration was achieved using epoch 4, batch size 16, and learning rate 5e-5 with the implementation of Focal Loss, resulting in a Macro F1-Score of 0.84. In addition, a web-based sentiment analysis system was developed to provide interactive visualization of prediction results. The findings demonstrate that optimizing the IndoBERT model with the Focal Loss approach is effective in providing accurate and in-depth analysis of digital banking service quality.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) > T58.6-58.62 Management Information Systems | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Firza Muhammad Pahlevi | ||||||||||||
| Date Deposited: | 26 May 2026 01:22 | ||||||||||||
| Last Modified: | 26 May 2026 01:22 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/52322 |
Actions (login required)
 |
View Item |