S. Irawan, D. Komariah (2025) PERBANDINGAN MODEL MACHINE LERANING UNTUK ANALISIS SENTIMEN KEPUASAN PENGGUNA APLIKASI AGEN PEGADAIAN. S1 thesis, Universitas Mercu Buana Jakarta - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (Cover)
41821010051-D. Komariah S. Irawan-01 Cover - D. Komariah S. Irawan.pdf Download (991kB) |
|
Text (BAB I)
41821010051-D. Komariah S. Irawan-02 Bab 1 - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (333kB) |
|
Text (BAB II)
41821010051-D. Komariah S. Irawan-03 Bab 2 - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (626kB) |
|
Text (BAB III)
41821010051-D. Komariah S. Irawan-04 Bab 3 - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (360kB) |
|
Text (BAB IV)
41821010051-D. Komariah S. Irawan-05 Bab 4 - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (1MB) |
|
Text (BAB V)
41821010051-D. Komariah S. Irawan-06 Bab 5 - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (281kB) |
|
Text (Daftar Pustaka)
41821010051-D. Komariah S. Irawan-07 Daftar Pustaka - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (447kB) |
|
Text (Lampiran)
41821010051-D. Komariah S. Irawan-08 Lampiran - D. Komariah S. Irawan.pdf Restricted to Registered users only Download (745kB) |
Abstract
Dalam era transformasi digital, Pegadaian mengembangkan aplikasi Agen Pegadaian guna meningkatkan aksesibilitas layanan keuangan bagi masyarakat serta memudahkan agen dalam bertransaksi dengan nasabah. Penelitian ini bertujuan untuk membandingkan kinerja dua model Machine Learning yaitu Support Vector Machine dan Naive Bayes dalam menganalisis sentimen kepuasan pengguna aplikasi Agen Pegadaian. Data yang digunakan dalam penelitian diperoleh melalui scraping ulasan pengguna aplikasi Agen Pegadaian di Google Play Store dengan jumlah data sebanyak 1200, yang kemudian diproses melalui tahapan preprocessing meliputi Cleaning, Case Folding, Normalization, Tokenizing, Stopword Removal dan Stemming. Setelah proses preprocessing selesai, dilakukan penghapusan baris kosong sehingga diperoleh sebanyak 922 data ulasan yang akan digunakan untuk proses pelabelan, pembagian data dan pengujian model. Hasil penelitian menunjukkan bahwa algoritma Naive Bayes menghasilkan akurasi yang sama pada pembagian data 90:10 dan 80:20 yaitu 89%. Sedangkan, algoritma Support Vector Machine menghasilkan akurasi 88% pada pembagian data 90:10 dan meningkat menjadi 89% pada pembagian data 80:20. Hasil dari penelitian ini diharapkan memberikan wawasan tentang model Machine Learning yang paling sesuai untuk menganalisis kepuasan pengguna aplikasi Agen Pegadaian, serta membantu pengambilan keputusan dalam pengembangan layanan aplikasi ke depannya. In the era of digital transformation, Pegadaian developed the Pegadaian Agent application to improve the accessibility of financial services for the public and facilitate agents in transacting with customers. This study aims to compare the performance of two Machine Learning models, namely Support Vector Machine and Naive Bayes in analyzing user satisfaction sentiments of the Pegadaian Agent application. The data used in the study were obtained by scraping user reviews of the Pegadaian Agent application on the Google Play Store with a total of 1200 data, which were then processed through preprocessing stages including Cleaning, Case Folding, Normalization, Tokenizing, Stopword Removal and Stemming. After the preprocessing process was completed, blank lines were removed to obtain 922 review data that would be used for the labeling process, data division and model testing. The results showed that the Naive Bayes algorithm produced the same accuracy for both 90:10 and 80:20 data splits, namely 89%. Meanwhile, the Support Vector Machine algorithm produced 88% accuracy for the 90:10 data split, increasing to 89% for the 80:20 data split.The results of this study are expected to provide insight into the most appropriate Machine Learning model for analyzing user satisfaction of the Pegadaian Agent application, as well as assisting decision- making in the development of future application services.
| Item Type: | Thesis (S1) |
|---|---|
| NIM/NIDN Creators: | 41821010051 |
| Uncontrolled Keywords: | Analisis Sentimen, Analisa Kepuasan, Agen Pegadaian, Machine Learning Sentiment Analysis, Satisfaction Analysis, Agen Pegadaian, Machine Learning |
| Subjects: | 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000.01-000.09 Standard Subdivisions of Computer Science, Information and General Works/Subdivisi Standar Dari Ilmu Komputer, Informasi, dan Karya Umum |
| Divisions: | Fakultas Ilmu Komputer > Sistem Informasi |
| Depositing User: | ZAIRA ELVISIA |
| Date Deposited: | 04 Sep 2025 01:50 |
| Last Modified: | 04 Sep 2025 01:50 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/97389 |
Actions (login required)
 |
View Item |
