2026/07/29 by S. V. Balshetwar, Dr. S. V. Balshetwar
Computer Science · #Advanced Malware Detection Techniques #Spam and Phishing Detection #User Authentication and Security Systems
paper · doi:10.1080/08874417.2026.2700519
openalex publication_date 2026/07/29 · openalex created_date 2026/07/30 · openalex updated_date 2026/07/31
The rapid growth of digital ecosystems and social media platforms has increased exposure to sophisticated cyber threats targeting critical infrastructures, including power grids, healthcare networks, and transportation systems. Traditional cybersecurity mechanisms often fail to detect or mitigate dynamic attacks such as bot-driven exploitation, disinformation campaigns, and real-time data breaches. This study aims to develop an AI-driven predictive threat intelligence framework to enhance threat detection, forecast attack vectors, and enable adaptive, real-time cybersecurity responses. The proposed methodology employs an AI-driven predictive threat intelligence framework integrating multi-source data from critical infrastructures and social media platforms. Techniques include the Spatial Collaborative Filtering Technique (SCFT) to identify spatial and behavioral threat correlations, Adaptive Predictive Analysis for forecasting attack vectors, and Cognitive Threat Mapping (CTM) to detect coordinated attacks and disinformation campaigns. Social Media Anomaly Detection (SMAD) monitors suspicious activities, while Real-Time Risk Recalibration (RTR) continuously updates security protocols, ensuring precise, adaptive, and timely cybersecurity across interconnected digital systems. The proposed methodology demonstrates superior performance with a detection accuracy of 92–94%, false positive rates of 5–6%, and an average detection time of 3.0–3.4 seconds. In comparison, traditional IDS/SIEM/firewall systems report 80–90% accuracy, 12–15% false positives, and detection times of 14–17.5 seconds. The proposed AI framework outperforms recent AI-based predictive solutions, such as the Adaptive Collaborative Intelligence (ACI) framework by Gupta, R. & Singh et al. by achieving up to 8% higher accuracy and reducing response latency by 60–70%. This indicates that the AI-driven framework enhances the resilience and proactivity of cybersecurity systems in both social media environments and critical infrastructures, offering promising capabilities in detecting phishing, ransomware, botnets, and cross-platform attacks. Future work will explore real-world deployment, the integration of explainable AI, multilingual threat detection, and self-evolving defense systems.