Deteksi Berita Online Hoax COVID-19 di Indonesia Menggunakan Metode Hybrid Long Short Term Memory dan Support Vector Machine

Anisa, Dwi Fitriaini Nur (2022) Deteksi Berita Online Hoax COVID-19 di Indonesia Menggunakan Metode Hybrid Long Short Term Memory dan Support Vector Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06111640000042-Undergraduate_Thesis.pdf] Text
06111640000042-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2024.
Download (2MB) | Request a copy
[thumbnail of 06111640000042-Undergraduate_Thesis.pdf] Text
06111640000042-Undergraduate_Thesis.pdf
Restricted to Repository staff only
Download (2MB) | Request a copy

Abstract

Fokus masyarakat Indonesia tidak lepas dari kasus pandemi Coronavirus Disease 2019 (COVID-19) dengan mengikuti setiap informasi terkait perkembangannya setiap hari. Hal ini yang mendorong banyak pihak terlebih pemerintah untuk menyediakan layanan informasi terkini terkait COVID-19. Namun, banyak berita online menyajikan informasi palsu yang dikenal dengan berita hoax tentang COVID-19 yang dapat menyebabkan keresahan masyarakat. Pada Tugas Akhir ini, dilakukan deteksi terhadap berita–berita online seputar informasi COVID-19 di Indonesia yang dibagi menjadi dua kategori, yaitu berita hoax dan berita fakta. Proses deteksi berita online dilakukan dengan metode penggabungan Long Short Term Memory dan Support Vector Machine (hybrid LSTM-SVM). LSTM menghasilkan fitur teks representatif yang selanjutnya digunakan untuk proses klasifikasi berita oleh SVM yang menghasilkan persentase nilai akurasi mencapai 94%.
================================================================================================
Most Indonesia's attention remains on COVID-19 cases by following the updated cases each day. In addition, many communities or, especially the Government make an effort to provide public services by delivering the recent COVID-19 situations news. Unfortunately, we still find numbers of hoax news about COVID-19 yielding massive public anxiety. Hence, this final project attempts to detect the online news about COVID-19 in Indonesia, whether hoax or fact news. In the detection process, we combined long short term memory and support vector machine, so-called hybrid LSTM-SVM. LSTM helps us to generate text representative features that classify them into hoaxes or facts using SVM which produce percentage of accuracy value reached 94%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: hoax news detection, COVID-19, LSTM, SVM, deteksi berita hoax
Subjects: Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Dwi Fitriaini Nur Anisa
Date Deposited: 04 Feb 2022 08:22
Last Modified: 04 Feb 2022 08:22
URI: http://repository.its.ac.id/id/eprint/92822

Actions (login required)

View Item View Item