Investigating Fake News Detection Using Machine Learning

Authors

  • Anshu Tuteja , Apurv Verma , Dr.Abhishek Badholia

Abstract

Due to increase in the online social network development in the past few years, due to different
political purposes and commercials fake news appear in large numbers and in the online world has a
widespread. By these online fake news online social networks users can get effected easily. Already in the
social media fake news has brought a remarkable impact. Fake news has truned into social problem in past
few years especially since the social media is rising and sometimes it spreads even faster and more than a
true information. We evaluate the performance in this paper for Fake news detection by Attention
mechanism on the dataset of two, one of which consists online traditional articles and other consists of
various sources news. Timely identification of fake news ia an important aim of improving online
information trustworthiness of the online social network. The goal of this paper is to investigate algorithms,
methodologies and principles for fake news subjects, creators ans articles detection from the online social
media and corresponding performance evaluations. Diverse connections along the subject, creator and news
article and Issues introduced by the unknown parameters of the fake news is addressed in this paper. A novel
automatic fake news credibility inference model is introduced in this research paper, known as FAKE
DETECTOR. On the basis of from the textual information, the set of latent and explicit features extraction, a
model of machine learning is built by FAKE DETECTOR, presentation of subject, creator and news articles
simultaneously. On real world fake news dataset extensible experiment is done to compare FAKE
DETECTOR with various models of state of art and demonstrating proposed model effectiveness by the
experimental results.

Published

2020-01-31

Issue

Section

Articles