Re-Fake: Classification of Fake Accounts on Online Social Media using the RNN Algorithm

  • Putra Wanda Prodi S-1 Informatika, Universitas Respati Yogyakarta, Indonesia
  • Marselina Endah Hiswati Prodi S-1 Informatika, Universitas Respati Yogyakarta, Indonesia
  • Mohammad Diqi Prodi S-1 Informatika, Universitas Respati Yogyakarta, Indonesia
  • Romana Herlinda Prodi S-1 Informatika, Universitas Respati Yogyakarta, Indonesia
Keywords: Classification, Fake Account, Recurrent Neural Network, Deep Learning

Abstract

Online Social Network (OSN) is an application that enables public communication and
information sharing. However, fake accounts on OSN can spread false information with unknown sources. It is a challenging task to detect malicious accounts in a large OSN system. The existence of fake accounts or unknown accounts on OSN can be a serious problem in maintaining data privacy. Various communities have proposed many techniques to deal with fake accounts on OSN, including rules-based black-and-white techniques to learning approaches. Therefore, in this study we propose a classification model using RNN to detect fake accounts accurately and effectively. We carried out this research in several steps, including collecting the dataset, pre-processing, extraction, training our model using RNN. Based on the experimental results, our proposed model can produce higher accuracy than conventional learning models.

Downloads

Download data is not yet available.
Published
2021-12-21
How to Cite
Wanda, P., Hiswati, M., Diqi, M., & Herlinda, R. (2021). Re-Fake: Classification of Fake Accounts on Online Social Media using the RNN Algorithm. Prosiding Seminar Nasional Sains Teknologi Dan Inovasi Indonesia (SENASTINDO), 3, 191 - 200. https://doi.org/10.54706/senastindo.v3.2021.139

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.