Fake News and Media Bias Detection: A Comparative Survey of Big Data Approaches

The widespread dissemination of fake news (FN) and misleading information across social media platforms has emerged as a critical challenge with profound social, political, and economic consequences. Since traditional content-based approaches often prove insufficient for identifying deceptive or biased information, researchers have increasingly shifted toward Big Data methodologies that simultaneously integrate linguistic features, user interactions, and network dynamics. This survey paper analyzes and compares three representative and prominent studies utilizing Big Data for FN and media bias detection. The first study generates a multidimensional dataset through crowdsourced annotation for various bias dimensions, such as subjectivity and framing. The second employs Geometric Deep Learning on social graphs to identify fake news based on propagation patterns within the network. Finally, the third work provides a comprehensive overview of data mining strategies, emphasizing the integration of content and social signals. We conduct a detailed analysis of the methodologies, datasets, and reported results, highlighting the pivotal contributions of these works to the evolving field of misinformation detection. The ultimate goal of this article is to provide a structured understanding of current Big Data approaches, identifying open challenges, future research directions, and the critical need for developing robust, scalable, and interpretable FN detection systems.

کلیدواژه ها:

Fake News ، Media Bias ، Big Data ، Geometric Deep Learning ، Social Networks ، Data Mining ، Misinformation Detection