Data-Driven Threat Scoring System for Dynamic Cyber Risk Management

Modern organizations operate in threat environments where cyber risks evolve faster than traditional assessment methods can respond. Static scoring frameworks often fail to capture real-time fluctuations in attack surfaces, adversary behavior, and contextual business impacts. This paper proposes a data-driven threat scoring system designed to support dynamic cyber risk management through continuous monitoring, adaptive analytics, and automated prioritization. The model integrates telemetry from network traffic, endpoint sensors, vulnerability databases, and threat intelligence feeds to generate a composite threat score that updates as conditions change. Machine learning techniques are employed to identify anomalies, predict potential exploitation paths, and weight risk factors according to their relevance in the current operational context. The system further incorporates business impact modeling to ensure that technical risk is aligned with organizational priorities. Experimental evaluation demonstrates that dynamic scoring improves detection sensitivity, shortens response time, and enhances overall situational awareness compared to static risk frameworks. The proposed system offers a scalable and practical approach for organizations seeking to transition toward proactive, intelligence-driven cyber risk management.