Towards Intelligent Enterprises: Adoption of AI for Cybersecurity Management and Risk Governance
Ravi Sekhar Kommuri, Mohanraju Muppala
Abstract
The modern digital business has a rapidly growing quantity of threats with automated and sophisticated cyber-attacks. Traditional, signature-based cyber security and fixed risk governance frameworks have been found to be more inadequate. The research paper argues that the shift to an Intelligent Enterprise relies on the strategic adoption of Artificial Intelligence (AI) and Machine Learning (ML) as the fundamental part of the cyber security management and risk governance. We introduce a new and progressive architectural design, which utilizes an unsupervised AI model that identifies anomalies and applies the supervised AI model to classify threats. An important innovation of our solution is context-based Query Fragment Caching Algorithm that has been developed to improve the functionality of real-time Security Information and Event Management (SIEM) systems. The algorithm can prioritize intelligently and cache query patterns that are computationally intensive and accessed frequently in an effort to reduce latency in threat analytics. The complexity analysis proves the presence of a logarithmic time complexity of a cache-retrieval operation. Experiments conducted on the NSL foundations and CIC foundations on NSL-KDD and CIC-IDS-2017 datasets indicate that the detection of threats has increased significantly and that the threat detector is now at 96.3 percent and F1-score of 94.1 percent whereas the mean query response time has decreased by 68 per cent compared to traditional Least-Recently-Used (LRU) caching systems. These results highlight the disruptive quality of AI in the creation of proactive, robust, and intelligent cyber-defence positions of contemporary companies.