Citation

Fahad, Nafiz and Goh, Michael Kah Ong and Hossen, Md. Ismail and Shahriar Shopnil, K. M. and Mitu, Israt Jahan and Alif, Md. A. Hossain and Tee, Connie (2023) Stand up Against Bad Intended News: An Approach to Detect Fake News using Machine Learning. Emerging Science Journal, 7 (4). pp. 1247-1259. ISSN 2610-9182

Text 1651-5503-1-PB.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The purpose of this approach is to find out the effects and efficiently detect fake news by using a publicly available dataset. However, it is difficult for human beings to judge an article's truthfulness manually, which is why This paper mainly wanted to cure the effect and to found out an automated fake news detection system with benchmark accuracy by using a machine learning classifier, which must be higher than other recent research works. In essence, this work’s target is to find out an efficient way to detect fake and real news, and it also the target is to compare with existing work where researchers used machine learning classifiers and deep learning architecture. The proposed approach depended on a systematic literature review and a publicly available dataset where 7796 news data are recorded with 50% real and 50% fake news. The best and benchmark accuracy is 93.61%, achieved by the Support Vector Machine (SVM) among the used Random Forest, Decision Tree, KNN, and Logistics Regression classifiers, and the achieved accuracy is better than the exciting recent research works. Moreover, fake news is detected, people are able to differentiate between fake or real news, and effects are cured when people used SVM.

Item Type:	Article
Uncontrolled Keywords:	Fake News; Rumor; False Information; Social Platforms; Machine Learning.
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	04 Sep 2023 01:57
Last Modified:	04 Sep 2023 01:57
URII:	http://shdl.mmu.edu.my/id/eprint/11638

Downloads

Edit (login required)