This study examined the effectiveness of predictive neural network models in improving cyberattack detection and vulnerability assessment in critical infrastructure systems, tackling the shortcomings of traditional machine learning methods in terms of accuracy, adaptability, and operational performance. Through a comprehensive review of peer-reviewed studies published between 2013 and 2023, the research synthesized recent advancements in machine learning, deep learning, and vulnerability analysis to create and evaluate an integrated predictive framework. The empirical analysis utilized a large-scale, real-world dataset comprising over 30 million network flow records, 12 million authentication and identity events, and more than 10,000 documented vulnerabilities from the energy, healthcare, and transportation sectors. The study employed convolutional neural networks (CNNs), gated recurrent units (GRUs), and hybrid CNN–GRU models, benchmarking them against logistic regression and random forest classifiers to assess improvements in detection accuracy, false positive reduction, vulnerability prioritization, and real-time performance. The vulnerability prioritization accuracy saw significant improvement, with around 25% increase in hit rates for the top 100 exploited vulnerabilities with real-world exploitation events. Latency and throughput metrics indicated that CNN detectors processed samples in under 2 milliseconds, while hybrid models handled event processing in less than 20 milliseconds, confirming their readiness for operational deployment. The study concludes that predictive neural network models represent a significant advancement in cybersecurity by effectively capturing nonlinear relationships, modelling IT–OT dependencies and integrating attack detection with vulnerability prioritization.
A Sun, study studied this question.