Multimodal Data Fusion Framework For Fake News Detection

Researchers have received considerable attention for automated fake news detection in social media in recent years. It can reduce time and space, aiding in reducing the manual effort. In this paper, we propose a deep learning-based multimodal fake news identification approach to detect fake news spread through social media. We used SpotFake+ as our base model and augmented the news article’s title information along with the article’s content and image. We compared the performance of our model with other existing models. The results obtained with the proposed model outperformed the existing methods in the literature. The model proposed has achieved an accuracy of 87% for the GossipCop dataset.