Febriany, Asri Kinanti (2023) Analisis Sentimen Berbasis Aspek Pada Ulasan Aplikasi InDrive Menggunakan Bidirectional Encoder From Transformers (BERT). Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text
cover.pdf Download (565kB) | Preview |
|
|
Text
BAB 1.pdf Download (47kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text
BAB 2.pdf Restricted to Registered users only until 31 December 2025. Download (222kB) | Request a copy |
|
|
Text
BAB 3.pdf Restricted to Registered users only until 31 December 2025. Download (199kB) | Request a copy |
|
|
Text
BAB 4.pdf Restricted to Registered users only until 19 December 2025. Download (2MB) | Request a copy |
|
|
Text
BAB 5.pdf Download (66kB) | Preview |
|
|
Text
DAFTAR PUSTAKA.pdf Download (86kB) | Preview |
|
|
Text
LAMPIRAN.pdf Restricted to Registered users only until 19 December 2025. Download (191kB) | Request a copy |
Abstract
The digital economy in Indonesia is significantly influenced by online transportation businesses, with inDrive emerging as a new player in this sector. Google Play Store ratings alone are insufficient to depict the true quality of an application. Hence, Aspect-Based Sentiment Analysis (ABSA) becomes crucial, as it can identify aspects and sentiments associated with them. Employing Latent Dirichlet Allocation (LDA), three main aspects were identified in the dataset from January to July: bidding features, application system, and transportation fares & services. The bidding feature's strengths lie in user convenience, although some drivers complain about the use of autobidding through bots or APK mods. The application system facilitates bookings, yet there are complaints about balance deductions and difficulty securing orders due to autobidding. Customers appreciate transportation fares & services, while drivers express dissatisfaction with low fare thresholds. Trend analysis reveals a balanced review of bidding features, with an increase in driver complaints in July. The application system dominates negative reviews but experiences a decline after improvements in March. Positive sentiments prevail in transportation fares & services, peaking in June but slightly decreasing in July. The ABSA model implementation using IndoBERT, validated with 100 new reviews, demonstrates success with 92% accuracy for aspect testing and 89% for sentiment testing.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76 Computer software Q Science > QA Mathematics > QA76.6 Computer Programming T Technology > T Technology (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Asri Kinanti Febriany | ||||||||||||
| Date Deposited: | 19 Dec 2023 07:51 | ||||||||||||
| Last Modified: | 19 Dec 2023 07:51 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/19000 |
Actions (login required)
 |
View Item |