Merinda, Siska (2025) PERBANDINGAN PERFORMA ALGORITMA SUPPORT VECTOR MACHINE DAN Bi-DIRECTIONAL LONG SHORT TERM MEMORY DALAM MENGKLASIFIKASI BERITA HOAKS. Diploma thesis, Politeknik Negeri Sriwijaya.
|
Text (Cover)
Cover.pdf - Published Version Download (653kB) | Preview |
|
|
Text (Abstrak)
ABSTRAK.pdf - Published Version Download (236kB) | Preview |
|
![[img]](http://eprints.polsri.ac.id/style/images/fileicons/text.png) |
Text (Bab I)
BAB I Pendahuluan.pdf - Published Version Restricted to Repository staff only Download (286kB) | Request a copy |
|
|
Text (Bab II)
BAB II Tinajauan Pustaka.pdf - Published Version Restricted to Repository staff only Download (523kB) | Request a copy |
|
|
Text (Bab III)
BAB III Metodologi Penelitian.pdf - Published Version Restricted to Repository staff only Download (545kB) | Request a copy |
|
|
Text (Bab IV)
BAB IV Hasil dan Pembahasan.pdf - Published Version Restricted to Repository staff only Download (774kB) | Request a copy |
|
|
Text (Bab V)
Bab V Penutup.pdf - Published Version Restricted to Repository staff only Download (221kB) | Request a copy |
|
|
Text (Daftar Pustaka)
Daftar Pustaka.pdf - Published Version Restricted to Repository staff only Download (191kB) | Request a copy |
|
|
Text (Lampiran)
Lampiran.pdf - Published Version Restricted to Repository staff only Download (2MB) | Request a copy |
|
|
Text (FULLTEXT)
FULLTEXT.pdf - Published Version Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Perkembangan teknologi digital dan kecerdasan buatan mendorong inovasi dalam deteksi otomatis berita hoaks. Penelitian ini membandingkan kinerja dua algoritma klasifikasi teks, yaitu Support Vector Machine (SVM) dan Bi-Directional Long Short-Term Memory (Bi-LSTM), untuk mendeteksi berita hoaks berbahasa Indonesia. Dataset terdiri dari 19.264 berita, terbagi rata antara hoaks dan non hoaks, dengan proporsi 80% data latih dan 20% data uji. Model SVM menggunakan pendekatan TF-IDF dan SVC linear, sementara Bi-LSTM menggunakan embedding layer dan dua lapisan LSTM dua arah. Evaluasi dilakukan menggunakan akurasi, precision, recall, dan F1-score. Hasilnya, SVM mencatat akurasi 98,70%, precision 99,26%, recall 98,13%, dan F1-score 98,68%. Bi-LSTM memberikan hasil lebih baik dengan akurasi 99,21%, precision 99,26%, recall 99,15%, dan F1-score 99,19%. Bi-LSTM terbukti lebih konsisten dalam mendeteksi hoaks. Kedua model telah diterapkan dalam sistem web untuk memudahkan masyarakat dalam mengidentifikasi berita hoaks secara cepat dan efisien.
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Berita Hoaks, Deteksi Hoaks, Evaluasi Model, Klasifikasi Teks, Sistem Berbasis Web, Support Vector Machine, Bi-Directional Long Short-Term Memory |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Telecommunication Engineering > Undergraduate Theses |
| Depositing User: | Pustaka Teknik Elektro |
| Date Deposited: | 29 Aug 2025 08:26 |
| Last Modified: | 29 Aug 2025 08:26 |
| URI: | http://eprints.polsri.ac.id/id/eprint/18951 |
Actions (login required)
 |
View Item |
