Multimodal Approaches based on Fake News Detection

People or organizations acting in their own or other people's best interests may produce fake news. Misinformation is typically produced with personal, political, or economic objectives in mind. Fake news is mainly known as one of the most severe dangers to worldwide commerce, and journalism with significant dependent team. To detect the fake news, many approaches are used. The proposed model used machine learning (ML) for exploiting the multi-domain visual data information in identification of fake news. In this work, an application considered to predict whether an image or video has been manipulated or not. And also, OCR is used for identifying the text on image and model will forecast outcome. The present system, used machine learning techniques like Naïve Bayes and support vector machine to predict the fake news. The present system achieved 97% accuracy to forecast model outcome.