Rasidi, Ahmad Nafi and -, Endang Wahyu Pamungkas, S.Kom., M.Kom., Ph.D. (2026) Perbandingan Varian Naive Bayes Classifier untuk Deteksi Berita Hoaks Berbahasa Indonesia pada Dataset Seimbang dan Tidak Seimbang. Skripsi thesis, Universitas Muhammadiyah Surakarta.
![[img]](https://eprints.ums.ac.id/style/images/fileicons/application_pdf.png) |
PDF (Surat Pernyataan Publikasi)
SURAT PERNYATAAN.pdf Restricted to Repository staff only Download (272kB) |
|
PDF (Naskah Publikasi)
NASKAH PUBLIKASI_revisi.pdf Download (1MB) |
Abstract
The rapid advancement of information technology has accelerated the spread of fake news (hoaxes) in Indonesia, causing public anxiety and social polarization. Manual factchecking is ineffective at large scale, highlighting the need for automated artificial intelligence solutions. This study compares the performance of three Naive Bayes classifiers—Multinomial, Bernoulli, and Gaussian—for detecting Indonesian-language hoaxes and examines the impact of dataset balancing. The research utilized 27,433 Indonesian news articles comprising both hoax and valid news. After preprocessing (case folding, tokenization, stopword removal, and stemming), TF-IDF was applied for feature extraction. The dataset was split into 80% training and 20% testing sets. Performance was evaluated using accuracy, precision, and recall based on the Confusion Matrix. Results show that dataset balancing significantly improves model performance. In the balanced scenario, Bernoulli Naive Bayes achieved the highest hoax recall (0.960) with 0.897 accuracy. Multinomial Naive Bayes obtained the best overall accuracy (0.942) and precision (0.959), while Gaussian Naive Bayes performed the lowest. This study concludes that Bernoulli Naive Bayes is more effective for Indonesian hoax detection than the other variants.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Uncontrolled Keywords: | Hoax detection, Naive Bayes Classifier, TF-IDF, fake news |
| Subjects: | T Technology > T Technology (General) T Technology > Information Technology |
| Divisions: | Fakultas Komunikasi dan Informatika > S1 Teknik Informatika |
| Depositing User: | AHMAD NAFI RASIDI |
| Date Deposited: | 16 May 2026 03:27 |
| Last Modified: | 16 May 2026 03:27 |
| URI: | http://eprints.ums.ac.id/id/eprint/145448 |
Actions (login required)
 |
View Item |
