Atmaja, Ferdy (2024) Analisis Sentimen Berbasis Aspek pada Sistem Layanan Pengaduan Masyarakat di Kota Surabaya Menggunakan Metode Latent Dirichlet Allocation dan Naïve Bayes. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (cover)
20082010118.-cover.pdf Download (3MB) |
|
Text (bab 1)
20082010118.-bab1.pdf Download (152kB) |
|
Text (bab 2)
20082010118.-bab2.pdf Restricted to Repository staff only until 16 December 2026. Download (353kB) | Request a copy |
|
Text (bab 3)
20082010118.-bab3.pdf Restricted to Repository staff only until 16 December 2026. Download (331kB) | Request a copy |
|
Text (bab 4)
20082010118.-bab4.pdf Restricted to Repository staff only until 16 December 2026. Download (2MB) | Request a copy |
|
Text (bab 5)
20082010118.-bab5.pdf Download (52kB) |
|
Text (daftar pustaka)
20082010118.-daftarpustaka.pdf Download (188kB) |
|
Text (lampiran)
20082010118.-lampiran.pdf Restricted to Repository staff only Download (198kB) | Request a copy |
Abstract
The public complaint service system is a crucial platform for city governments to receive, manage, and respond to citizens' grievances. However, the increasing volume of complaints in Surabaya poses challenges for manual management, which is often inefficient. In-depth analysis of complaint data is essential to understand public opinions regarding service quality. This study aims to perform aspect-based sentiment analysis on public complaint data using Latent Dirichlet Allocation (LDA) and Naïve Bayes methods. LDA is employed to identify key aspects frequently mentioned in complaints, such as services, infrastructure, and cleanliness, while Naïve Bayes is applied to classify public sentiment into positive, negative, or neutral categories. The dataset consists of 10,847 complaint records collected through the WargaKu application and the Media Center of Surabaya City throughout 2023. The findings reveal that LDA successfully identified 17 key topics, including administration, infrastructure, and public information, while the Naïve Bayes model, enhanced with resampling techniques, achieved an accuracy of 80% in sentiment classification. This study provides data-driven insights to help city governments prioritize improvements on critical issues and offers an approach that can be adopted by similar complaint management systems.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76.6 Computer Programming T Technology > T Technology (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Ferdy Atmaja | ||||||||||||
| Date Deposited: | 16 Dec 2024 05:48 | ||||||||||||
| Last Modified: | 16 Dec 2024 05:48 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/33434 |
Actions (login required)
 |
View Item |