Coupled Matrix–Tensor Factorization (Cmtf) And Reinforcement Deep Belief Network (Rdbn) For Fake News Detection In Social Media
Year: 2025
Type: article
Abstract: Social media's quick information-sharing capabilities have made it a popular platform for individuals to interact with one another and exchange ideas. Artificial Intelligence (AI) and Machine Learning (ML) have been developed for fake news detection on social media, thereby enhancing people's daily lives. It is essential to remove irrelevant features from social media to increase detection accuracy of fake news. High dimensional datasets including content, context, and community-level aspects cannot be handled by effective detection models. In this paper, a novel two-step method has been developed on social media for fake news detection. Initial step, Coupled Matrix–Tensor Factorization (CMTF) method is used to tensor formation. When dealing with labeled data, the class information with factorization procedure is introduced for fake news detection. Second step, Reinforcement Deep Belief Network (RDBN) model is developed for fake news detection. Reinforcement Restricted Boltzmann Machine (RRBM) is created by incorporating the reinforcement learning idea into the trained RBM and Back Propagation (BP) technique for label attachment. Finally the performance of the detection methods has been validated using BuzzFeed and PolitiFact in terms of Precision (P), Recall (R), F1-score (F1), and Accuracy (A)
Source: Journal of Neonatal Surgery
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7219375643
