IMPLEMENTING XLNET SYSTEM AND CONTEXTUAL DATA AUGMENTATION FOR HOAX NEWS DETECTION ON FACEBOOK SOCIAL MEDIA

SUNU, BAYU ANGGORO (2026) IMPLEMENTING XLNET SYSTEM AND CONTEXTUAL DATA AUGMENTATION FOR HOAX NEWS DETECTION ON FACEBOOK SOCIAL MEDIA. Tugas Akhir thesis, University of Technology Yogyakarta.

[img] Text
5220411120_Bayu Anggoro Sunu_Abstrak.pdf
Download (134kB)

Abstract

ABSTRACT The increasing spread of hoax news on social media, particularly Facebook, has serious consequences, including public misunderstanding, declining information quality, and the potential for social conflict. Data from the Ministry of Communication and Informatics (Kemenkominfo) in 2023 shows that of the 526 hoaxes identified, 455 were disseminated via Facebook, highlighting the need for more effective countermeasures. A significant challenge is the absence of an automated verification system that enables the public to quickly and independently assess the veracity of information. This research aims to develop a web-based hoax news detection system to serve as a verification tool before information is disseminated. The system was built using the XLNet algorithm, which has strong capabilities for deep language understanding, and incorporates data augmentation techniques such as synonym replacement, back translation, and paraphrasing to enhance the quality and diversity of the dataset. The results demonstrate that the system can accurately classify news as hoax or valid while providing a responsive user experience via a simple, interactive web interface. The system analyses user-entered news text and delivers real-time classification results to support efforts to reduce the spread of hoaxes on social media. Keywords: Hoax Detection, Social Media, XLNet, Data Augmentation, Natural Language Processing, Facebook.

Item Type: Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects: T Technology > T Technology (General)
T Technology > T Technology (General) > T201 Patents. Trademarks
Divisions: Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User: Kaprodi S1 Informatika UTY
Date Deposited: 06 May 2026 04:09
Last Modified: 06 May 2026 04:09
URI: http://eprints.uty.ac.id/id/eprint/19796

Actions (login required)

View Item View Item
UTY Open Access is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.