CyberInfo: A Comprehensive Review of AI-Driven Cyber Threat Intelligence, LLM Based Analysis, MITRE ATT&CK Mapping, and Incident Correlation Systems
Abstract
The increasing complexity of cyberattacks has created a growing demand for intelligent Cyber Threat Intelligence (CTI) systems capable of processing large volumes of security data and generating actionable insights. Traditional threat analysis methods rely heavily on manual investigation, resulting in delayed detection and response to emerging threats. Recent advancements in Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG) have significantly enhanced the automation of threat intelligence workflows. This review paper examines existing research on AI-driven cyber threat intelligence, automated threat extraction, MITRE ATT&CK mapping, anomaly detection, and incident correlation techniques. Various methodologies, architectures, and frameworks proposed by researchers are analyzed and compared to identify their strengths, limitations, and practical applications. The study further highlights existing research gaps, including limited integration of intelligence sources, lack of contextual threat correlation, and challenges in real-time threat analysis. Based on the reviewed literature, the paper discusses the development of CyberInfo, an AI-powered threat intelligence and incident correlation platform designed to improve threat visibility, reduce analyst workload, and support proactive cybersecurity operations