Almira Darma Utami Fadillah, - (2025) ANALISIS SENTIMEN ULASAN APLIKASI GOJEK MENGGUNAKAN ALGORITMA KOMBINASI K-MEANS DAN RANDOM FOREST. S1 thesis, Universitas Pendidikan Indonesia.
Abstract
Aplikasi transportasi online seperti Gojek telah merubah cara masyrakat menggunakan layanan transportasi, dengan menawarkan berbagai fitur tambahan seperti pemesanan makanan dan pengiriman barang. Pengguna dapat memberikan ulasan yang mencerminkan pengalaman mereka dalam menggunakan aplikasi, namun sering kali terdapat ketidaksesuaian antara rating dan ulasan yang diberikan. Oleh karena itu, analisis sentimen diperlukan untuk memahami opini, emosi, dan sikap yang terkandung dalam teks ulasan. Penelitian ini menggabungkan algoritma K-Means untuk klasterisasi dan Random Forest untuk klasifikasi. Proses analisis sentimen ini menggunakan data sebanyak 121.815 yang diambil dari situs Kaggle dari bulan November tahun 2021 hingga Februari tahun 2024. Hasil dari proses clustering mendapatkan 4 cluster, lalu hasil evaluasi dari penggabungan algortima K-Means dan Random Forest mendapatkan nilai akurasi sebesar 99%. Penelitian analisis sentimen ini diharapkan dapat memberikan informasi penting bagi pengembang aplikasi Gojek, terutama dalam memahami pola sentimen pengguna dan mengidentifikasi area yang memerlukan perbaikan dan penggunaan kombinasi algoritma tersebut diharapkan dapat meningkatkan nilai accuracy. ------------- Online transportation applications such as Gojek have changed the way people use transportation services by offering various additional features such as food ordering and goods delivery. Users can provide ratings that reflect their experience in using the application, but there is often a mismatch between the rating and the rating given. Therefore, sentiment analysis is needed to understand the opinions, emotions, and attitudes contained in the review text. This research combines KMeans algorithm for clustering and Random Forest for classification. This sentiment analysis process uses 121,815 data collected from the Kaggle website from November 2021 to February 2024. The results of the clustering process get 4 clusters, then the evaluation results from combining the K-Means and Random Forest algorithms get an accuracy value of 99%. This sentiment analysis research is expected to provide important information for Gojek application developers, especially in understanding user sentiment patterns and identifying areas that need improvement, and the use of a combination of these algorithms is expected to increase the accuracy value
![[img]](http://repository.upi.edu/style/images/fileicons/text.png) |
Text
S_RPL_2003354_Title.pdf Download (487kB) |
|
Text
S_RPL_2003354_Chapter1.pdf Download (105kB) |
|
Text
S_RPL_2003354_Chapter2.pdf Restricted to Staf Perpustakaan Download (314kB) |
|
Text
S_RPL_2003354_Chapter3.pdf Download (284kB) |
|
Text
S_RPL_2003354_Chapter4.pdf Restricted to Staf Perpustakaan Download (296kB) |
|
Text
S_RPL_2003354_Chapter5.pdf Download (31kB) |
|
Text
S_RPL_2003354_Appendix.pdf Restricted to Staf Perpustakaan Download (274kB) |
| Item Type: | Thesis (S1) |
|---|---|
| Uncontrolled Keywords: | Gojek, Analisis Sentimen, K-Means, Random Forest |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) T Technology > TN Mining engineering. Metallurgy |
| Divisions: | UPI Kampus cibiru > S1 Rekayasa Perangkaat Lunak |
| Depositing User: | Almira Darma Utami Fadillah |
| Date Deposited: | 05 Mar 2025 04:29 |
| Last Modified: | 05 Mar 2025 04:29 |
| URI: | http://repository.upi.edu/id/eprint/130162 |
Actions (login required)
 |
View Item |