FAHMI, SYAHRUL (2025) PERBANDINGAN ALGORITMA NAIVE BAYES, RANDOM FOREST, DAN SUPPORT VECTOR MACHINE PADA ANALISIS SENTIMEN ULASAN APLIKASI SEKOLAH.MU DI GOOGLE PLAY STORE. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 Cover.pdf Download (514kB) | Preview |
|
![]() |
Text (BAB I)
02 Bab 1.pdf Restricted to Registered users only Download (88kB) |
|
![]() |
Text (BAB II)
03 Bab 2.pdf Restricted to Registered users only Download (113kB) |
|
![]() |
Text (BAB III)
04 Bab 3.pdf Restricted to Registered users only Download (97kB) |
|
![]() |
Text (BAB IV)
05 Bab 4.pdf Restricted to Registered users only Download (916kB) |
|
![]() |
Text (BAB V)
06 Bab 5.pdf Restricted to Registered users only Download (26kB) |
|
![]() |
Text (DAFTAR PUSTAKA)
07 Daftar Pustaka.pdf Restricted to Registered users only Download (133kB) |
|
![]() |
Text (LAMPIRAN)
08 Lampiran.pdf Restricted to Registered users only Download (559kB) |
Abstract
Sentiment analysis of application reviews is essential for developers to understand user perceptions of their products. This study aims to compare the performance of three machine learning algorithms Naïve Bayes, Random Forest, and Support Vector Machine (SVM) in sentiment analysis of Sekolah.mu application reviews on Google Play Store. The research utilizes a dataset consisting of reviews with both automated and manual labeling, and applies the SMOTE (Synthetic Minority Oversampling Technique) to address data imbalance. The results show that implementing SMOTE generally improves model accuracy, particularly for datasets with automated labeling. In the 90:10 training-to-testing ratio scenario, SVM with SMOTE achieved the highest accuracy of 95.9% for automated labeling, while Random Forest provided the best results for manual labeling with an accuracy of 95.9% without SMOTE. Naïve Bayes exhibited significant accuracy improvement after applying SMOTE, especially for automated labeling, achieving 91.6% accuracy in the 90:10 ratio scenario. Overall, Random Forest and SVM outperformed Naïve Bayes, particularly in manual labeling scenarios. This study demonstrates that selecting the appropriate algorithm and data balancing techniques can enhance sentiment analysis performance, providing valuable insights for application developers to better understand user reviews. Keywords : Sentiment analysis, Naive Bayes, Support Vector Machine, Random Forest, Google Play Store, Sekolah.mu Analisis sentimen ulasan aplikasi menjadi penting bagi pengembang dalam memahami persepsi pengguna terhadap produk mereka. Penelitian ini bertujuan membandingkan kinerja tiga algoritma pembelajaran mesin, yaitu Naïve Bayes, Random Forest, dan Support Vector Machine (SVM), dalam analisis sentimen ulasan aplikasi Sekolah.mu di Google Play Store. Penelitian menggunakan dataset yang terdiri dari ulasan dengan pelabelan otomatis dan manual, serta menerapkan teknik SMOTE (Synthetic Minority Oversampling Technique) untuk mengatasi ketidakseimbangan data. Hasil pengujian menunjukkan bahwa penerapan SMOTE secara umum meningkatkan akurasi model, terutama untuk dataset dengan pelabelan otomatis. Pada skenario rasio data latih dan uji 90:10, algoritma SVM dengan SMOTE mencapai akurasi tertinggi sebesar 95,9% untuk pelabelan otomatis, sementara Random Forest memberikan hasil terbaik pada pelabelan manual dengan akurasi 95,9% tanpa SMOTE. Naive Bayes menunjukkan peningkatan akurasi signifikan setelah penerapan SMOTE, terutama pada pelabelan otomatis, dengan hasil mencapai 91,6% pada rasio data 90:10. Secara keseluruhan, Random Forest dan SVM menunjukkan performa yang lebih unggul dibandingkan Naïve Bayes, terutama dalam skenario pelabelan manual. Penelitian ini membuktikan bahwa pemilihan algoritma dan teknik penyeimbangan data yang tepat dapat meningkatkan kinerja analisis sentimen, serta memberikan wawasan bagi pengembang aplikasi dalam memahami ulasan pengguna secara lebih efektif. Kata Kunci : Analisis sentimen, Naive Bayes, Support Vector Machine, Random Forest, Google Play Store, Sekolah.mu
Actions (login required)
![]() |
View Item |