<![CDATA[The rapid growth of digital infrastructures has amplified cybersecurity vulnerabilities, challenging organizations to manage risks effectively. Traditional vulnerability assessment methods, such as static scoring systems, often overlook dynamic threat information, leading to suboptimal prioritization. This study addresses the gap in existing vulnerability management approaches by introducing a data-driven framework that combines internal system data, public vulnerability databases, and external threat intelligence using predictive analytics. The proposed decision support information system employs machine learning as an analytical component to estimate the likelihood of vulnerability exploitation and support vulnerability prioritization decisions. The novelty of this approach lies in its ability to prioritize vulnerabilities not only based on technical severity but also considering the context of real-world threat activity. When benchmarked against conventional methods, this approach demonstrates superior performance in identifying exploitable vulnerabilities, improving accuracy and recall, thus optimizing resource allocation. By adopting a proactive, risk-based strategy, the framework prioritizes the most critical vulnerabilities in complex IT environments. The results highlight the potential of predictive models in enhancing cybersecurity management and supporting sustainable infrastructure, driving a shift toward more efficient, data-driven decision-making. ]]>
Copyrights © 2026
