GRAPH-BASED BIG DATA ANALYTICS FOR REAL-TIME DETECTION OF COORDINATED MISINFORMATION CAMPAIGNS

Abstract

The rapid growth of social media platforms has amplified the spread of misinformation, often through coordinated campaigns involving networks of automated accounts and malicious actors. Traditional content-based detection methods struggle to identify such campaigns in real time due to their dynamic and networkdriven nature. This paper proposes a GraphBased Big Data Analytics Framework for the real-time detection of coordinated misinformation campaigns. The proposed system models social media interactions as dynamic graphs, where nodes represent users or posts and edges represent interactions such as shares, mentions, or retweets. By leveraging graph analytics techniques combined with scalable big data processing frameworks, the system detects anomalous community structures, synchronized behavior patterns, and information propagation anomalies. Graph-based metrics such as centrality, clustering coefficient, and community detection are integrated with machine learning models to identify coordinated inauthentic behavior. The framework is designed for high-throughput stream processing to ensure timely detection in large-scale social networks. Experimental evaluation demonstrates improved detection accuracy and reduced response time compared to traditional content-based approaches. The proposed model provides a scalable, real-time solution for mitigating misinformation threats in digital ecosystems.