PUTRI, ASAFITA DWI (2025) ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP RENCANA KENAIKAN PPN 12% DI INDONESIA DENGAN NAIVE BAYES, KNN, DAN RANDOM FOREST. S1 thesis, Universitas Mercu Buana Jakarta - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (COVER)
41820120038-Asafita Dwi Putri-01 Cover - Fita.pdf Download (326kB) |
|
Text (BAB I)
41820120038-Asafita Dwi Putri-02 Bab 1 - Fita.pdf Restricted to Registered users only Download (106kB) |
|
Text (BAB II)
41820120038-Asafita Dwi Putri-03 Bab 2 - Fita.pdf Restricted to Registered users only Download (226kB) |
|
Text (BAB III)
41820120038-Asafita Dwi Putri-04 Bab 3 - Fita.pdf Restricted to Registered users only Download (132kB) |
|
Text (BAB IV)
41820120038-Asafita Dwi Putri-05 Bab 4 - Fita.pdf Restricted to Registered users only Download (874kB) |
|
Text (BAB V)
41820120038-Asafita Dwi Putri-06 Bab 5 - Fita.pdf Restricted to Registered users only Download (297kB) |
|
Text (DAFTAR PUSTAKA)
41820120038-Asafita Dwi Putri-08 Daftar Pustaka - Fita.pdf Restricted to Registered users only Download (312kB) |
|
Text (LAMPIRAN)
41820120038-Asafita Dwi Putri-09 Lampiran - Fita.pdf Restricted to Registered users only Download (302kB) |
Abstract
Penelitian ini membahas analisis sentimen pengguna Twitter terkait rencana kenaikan Pajak Pertambahan Nilai (PPN) 12% di Indonesia dengan metode machine learning. Sebanyak 2.985 tweet dikumpulkan menggunakan kata kunci terkait, kemudian diproses melalui tahapan preprocessing dan pelabelan otomatis memakai RoBERTa. Distribusi data awal sangat tidak seimbang, didominasi sentimen netral (53%), negatif (35%), dan positif (12%). Tiga algoritma yakni Naive Bayes, K-Nearest Neighbor (KNN), dan Random Forest diuji sebelum dan sesudah penyeimbangan data dengan SMOTE. Setelah dilakukan penyeimbangan data menggunakan SMOTE, model Naive Bayes mencapai akurasi sebesar 88%, precision 88%, recall 88%, dan f1-score 87%. Model KNN memperoleh akurasi 65%, precision 75%, recall 65%, dan f1-score 56%. Sementara itu, model Random Forest menghasilkan hasil terbaik dengan akurasi 89%, precision 90%, recall 89%, dan f1-score 89%. This study discusses sentiment analysis of Twitter users regarding the planned 12% Value Added Tax (VAT) increase in Indonesia using machine learning methods. A total of 2,985 tweets were collected using relevant keywords, then processed through a series of preprocessing steps and automatically labeled using RoBERTa. The initial data distribution was highly imbalanced, dominated by neutral sentiment (53%), negative (35%), and positive (12%). Three algorithms— Naive Bayes, K-Nearest Neighbor (KNN), and Random Forest—were tested before and after data balancing with SMOTE. After balancing the data using SMOTE, the Naive Bayes model achieved an accuracy of 88%, precision of 88%, recall of 88%, and an f1-score of 87%. The KNN model obtained an accuracy of 65%, precision of 75%, recall of 65%, and an f1-score of 56%. Meanwhile, the Random Forest model produced the best results, with an accuracy of 89%, precision of 90%, recall of 89%, and an f1-score of 89%.
| Item Type: | Thesis (S1) |
|---|---|
| NIM/NIDN Creators: | 41820120038 |
| Uncontrolled Keywords: | Analisis Sentimen, Text Mining, Media Sosial, RoBERTa, SMOTE Sentiment Analysis, Text Mining, Social Media, RoBERTa, SMOTE |
| 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 > 003 Systems/Sistem-sistem |
| Divisions: | Fakultas Ilmu Komputer > Sistem Informasi |
| Depositing User: | NAIMAH NUR ISLAMIDIYANAH |
| Date Deposited: | 26 Aug 2025 02:31 |
| Last Modified: | 26 Aug 2025 02:31 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/97098 |
Actions (login required)
 |
View Item |
