Using Emerging Deep Convolutional Neural Networks (DCNN) Learning Techniques for Detecting Phony News

Abstract

Over the years, social media has revolutionized the way people share and receive information. The rapid dissemination of false information is another concern that may have negative consequences for individuals and society as a whole. For several economic and political reasons, fake news has started appearing online often and in massive amounts. One of the many stylistic tactics used by fake news producers to make their articles more appealing is appealing to readers' emotions. One of the many stylistic tactics used by fake news producers to make their articles more appealing is to appeal to readers' emotions. This has made it very challenging to identify fake news stories and help their producers validate them via data processing channels without deceiving the audience. Claims, particularly those that gain thousands of views and likes before being challenged and debunked by credible sources, need a method for fact-checking. In order to properly detect and classify fake news, many machine learning techniques have been implemented. In this experiment, an ML classifier was employed to ascertain the veracity of news reports. The best features of the dataset are used to evaluate the proposed model in comparison to other benchmark approaches. Our proposed model (DCNNs) outperforms the state-of-the-art methods in terms of classification accuracy (99.23 percent).