Deep Learning Driven Anomaly Detection in Social Graphs  by Using Anti-Corona Political Optimization

Anomaly detection in social networks is critical to identify abnormal behavior and prevent looming ‎security attacks. Anti-Corona Political Optimization (ACPO) is a novel anomaly detection ‎approach in social networks, utilizing advanced optimization techniques and deep learning ‎models for high-precision anomaly detection. ACPO integrates Anti-Corona Virus Optimization ‎‎(ACVO) and Political Optimization (PO) for efficient anomaly detection. The approach involves ‎pre-processing to deal with missing values, noise, and inconsistencies; followed by taking features ‎through the methodology of feature selection and data augmentation for enhanced robustness of the ‎model, followed by clustering to find patterns and anomaly detection in discovered clusters. ‎A combination of ACVO and PO enhances the detection of anomalies in ACPO; then deep learning ‎architecture will provide strong and accurate detection. In general, ACPO offers a novel and ‎effective approach with widespread applications in social network security and other fields. ‎Experimental results verify the improved performance of ACPO_DRN, which achieved precision ‎‎(0.813), recall (0.880), and F-measure (0.836) with K-Fold variations. The good performance of ‎ACPO in social network anomaly detection is demonstrated by its ability to learn complex patterns ‎and anomalies, and therefore, it is an effective tool for preventing possible security threats‎.