An Energy-Efficient Ensemble-Based Computational Social System for Fake News Detection in MANET Messaging

Mobile Adhoc Networks (MANETs) are created by the interconnection of energy-constrained mobile devices. One of the issues that MANETs face is the detection of fake news in messages sent over the network. To address the issue of fake news detection in an energy-constrained setting on mobile devices, this work proposes the Veracity architecture, an Energy-Efficient Ensemble-Based Computational Social System for fake news detection in MANET messaging. Veracity utilises five algorithms namely VerifyNews, CompareText, PredictCred, CredScore and EyeTruth for the prediction of the validity of the news. Veracity employs the Legitimacy ensemble model for the prediction of the quality of the news. The experimental energy efficiency results demonstrate that Veracity uses an ensemble model and negligibly affects the residual energy capacity of each device in the network.