Rohman, Muhammad Nur Afiga (2026) Implementasi Sistem Klasifikasi Sentimen Ulasan Pengguna Aplikasi Access by KAI Menggunakan Model IndoBERT. S1 thesis, Universitas PGRI Madiun.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
halaman depan.pdf Download (1MB) |
|
Text
abstrak.pdf Download (11kB) |
|
Text
bab1.pdf Download (142kB) |
|
Text
bab2.pdf Download (833kB) |
|
Text
bab3.pdf Restricted to Repository staff only Download (95kB) |
|
Text
bab4.pdf Restricted to Repository staff only Download (3MB) |
|
Text
bab5.pdf Download (14kB) |
|
Text
dafpus.pdf Download (282kB) |
|
Text
lampiran.pdf Restricted to Repository staff only Download (1MB) |
Abstract
Muhammad Nur Afiga Rohman. 2026. Implementation of a Sentimen Classification System for User Reviews of the Access by KAI Application Using the IndoBERT Model. Thesis. Informatics Engineering Study Program, Faculty of Engineering, Universitas PGRI Madiun. Advisor (I) Yessi Yunitasari, S.Kom., M.Cs., Advisor (II) Puguh Jayadi, S.Kom., M.Kom. User reviews of the Access by KAI application on the Google Play Store are available in large quantities and consist of unstructured text containing positive and netral responses, complaints, criticisms, and suggestions. This condition causes manual analysis to require considerable time and may lead to differences in interpretation. To address this problem, a system is needed to automatically classify the sentiment and aspects of User reviews and present the results in an understandable form. This study aims to design, develop, and implement a website-based sentiment classification system for User reviews of the Access by KAI application using the IndoBERT model. The system development method used in this study is the Prototype method, which consists of requirements analysis, prototype development, prototype evaluation, system development, testing, and system implementation. The results show that the system was successfully implemented using Vue 3 and TypeScript as the frontend, FastAPI as the backend, PostgreSQL as the database, and ONNX Runtime to execute the IndoBERT model. Review data are collected from the Google Play Store through a scraping process and then processed using IndoBERT to classify the reviews into negative, neutral, or positive sentimens, while aspect categories are determined using Rule Based Classification. The analysis results are stored in the database and presented through the Dashboard, Live Scraping, AI Pipeline, and Review Feedback features. Testing using Black Box Testing and a Confusion Matrix shows that the main system functions operate according to the specified requirements and that the model can be applied to support the sentiment classification process. Therefore, the system can assist in collecting, processing, storing, and presenting information from Access by KAI User reviews in a more structured manner
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | Kata Kunci: Access by KAI, Analisis Sentimen, IndoBERT, Natural Language Processing, Sistem Berbasis Web. Keywords: Access by KAI, IndoBERT, Natural Language Processing, Sentiment Analysis, Web-Based System. |
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Teknik > Teknik Informatika |
| Depositing User: | AFIGA NUR MUHAMMAD |
| Date Deposited: | 18 Aug 2026 07:13 |
| Last Modified: | 18 Aug 2026 07:13 |
| URI: | http://eprint.unipma.ac.id/id/eprint/9197 |
Actions (login required)
 |
View Item |
