CSSLnO: Cat Swarm Sea Lion Optimization-based deep learning for fake news detection from social media

Social media has effectively shortened the time for the distribution of information, which sometimes carry news when compared to traditional methods. The convenience and affordable instant access to data with revolution in mobile technology have directed the production of false news. Fake news has the potential to mobilize public opinion, causing social unrest. Therefore, it is essential to check the authenticity and credibility of news articles being shared on social media. Moreover, versions of fake news are very similar to the actual ones, so it is problematic for humans to recognize them. Therefore, Fake news identification from social media become an Artificial Intelligence (AI) problem. A Cat Swarm Sea Lion Optimization (CSSLnO)-based Deep Quantum Neural Network (DQNN) is developed in this research for detecting fake news. The input is effectively acquired from different websites, such as Twitter and Facebook. Various unique features are extracted from FakeNewsNet and BuzzFeed-Webis Fake News Corpus for the effective fake news detection process. DQNN is applied for identifying fake news using the extracted features. The authors developed an optimization algorithm named CSSLnO and employed it to further train the DQNN model to increase the detection performance. The developed model outperformed the other existing fake news detection techniques with improved accuracy, sensitivity, specificity, and F-measure of 90.65, 91.90, 89.09, and 91.46%.