MCNNet: Generalizing Fake News Detection with a Multichannel Convolutional Neural Network using a Novel COVID-19 Dataset

Abstract

During the pandemic of COVID-19, the propagation of fake news is spreading like wildfire on social media. Such fake news articles have created confusion among people and serious social disrup tions as well. To detect such news articles effectively, we propose a generalized classification model (MCNNet) having the power of learning across different kernel-sized convolutional layers in differ ent parallel channel network. The capability of MCNNet is lucrative towards any real-world fake news dataset. Experimental results have demonstrated the performance of our model with different real-world fake news datasets.