Lilmumazzaini, Farista (2026) ANALISIS SENTIMEN DAN KEPUASAN PENGGUNA APLIKASI E-GROCERY BERDASARKAN ULASAN GOOGLE PLAY STORE MENGGUNAKAN ALGORITMA NAIVE BAYES DAN XGBOOST. Undergraduate thesis, Universitas Pembangunan Nasional "Veteran" Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22032010088-cover.pdf Download (1MB) |
|
Text (Bab 1)
22032010088-bab1.pdf Download (402kB) |
|
Text (Bab 2)
22032010088-bab2.pdf Restricted to Repository staff only until 12 October 2029. Download (501kB) |
|
Text (Bab 3)
22032010088-bab3.pdf Restricted to Repository staff only until 12 October 2029. Download (357kB) |
|
Text (Bab 4)
22032010088-bab4.pdf Restricted to Repository staff only until 12 October 2029. Download (5MB) |
|
Text (Bab 5)
22032010088-bab5.pdf Download (282kB) |
|
Text (Daftar Pustaka)
22032010088-daftarpustaka.pdf Download (219kB) |
|
Text (Lampiran)
22032010088-lampiran.pdf Restricted to Repository staff only until 12 October 2029. Download (6MB) |
Abstract
The rapid growth of e-grocery services in Indonesia demands continuous improvement in e-service quality and user experience. This study aims to analyze sentiment characteristics, compare the performance of Naive Bayes and XGBoost algorithms, and measure user satisfaction across Alfagift, Klik Indomaret, and HappyFresh application based on Google Play Store reviews. Labeling results revealed a negative sentiment dominance (56%), where Klik Indomaret recorded the highest dissatisfaction (64.4% negative), followed by Alfagift (56.4% negative), while HappyFresh achieved the highest satisfaction (52.8% positive). In the model evaluation and McNemar's Test, XGBoost significantly outperformed Naive Bayes (p-value < 0,05), achieving an accuracy of 86%, compared to Naive Bayes with 75% accuracy. It is recommended for Alfagift to focus on real-time stock synchronization optimization of instant couriers, and provision of self-cancellation features; Klik Indomaret should prioritize improvements to the OTP login system, cessation of unilateral order cancellations, as well as acceleration of refund SLA and customer support; whereas HappyFresh provides features for product substitution preferences, service fee transparency, and refunds to the original payment method.
| Item Type: | Thesis (Undergraduate) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||
| Subjects: | H Social Sciences > HD Industries. Land use. Labor > HD30.213 Decision Support Systems T Technology > T Technology (General) > T55.4-60.8 Industrial engineering. Management engineering T Technology > T Technology (General) > T58.6-58.62 Management Information Systems |
||||||||
| Divisions: | Faculty of Engineering > Departement of Industrial Engineering | ||||||||
| Depositing User: | Farista Lilmumazzaini | ||||||||
| Date Deposited: | 13 Oct 2026 03:55 | ||||||||
| Last Modified: | 14 Oct 2026 01:59 | ||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/60715 |
Actions (login required)
 |
View Item |