Fake News Classification Bimodal using Convolutional Neural Network and Long Short-Term Memory
Year: 2020
Type: preprint
Abstract: Fake news is a publicity or conspiracy that contains cautious or fake information that has a social as well as political impact because it is spread through old fashioned media and gets the progression via social or news media. In this article we are using multimodal approach with Convolutional Neural Network (CNN) and Long-Short term memory (LSTM) to classify the fake and fake news articles with high accuracy. We are working on database with 12 different categories of news articles. We used linguistic cue approaches with machine learning. We classified a news source based on its source and their previous history (such as Domain name or Author name) with bimodal CNN and LSTM. Through reputable news source our model classified post reliable news articles with the accuracy 99.7 % on the training data and 97.5 % for test data. But still we have to consider some fake news can still be published on reputable domain so it's better to use some other parameter such as news headline.
Source: SSRN Electronic Journal
Authors: Abdullah Abdullah, Mazhar Javed Awan, Hafiz Muhammad Faisal Shehzad, Muhammad Adnan Ashraf
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3198530906
