What does the prediction score mean?

Orpheus predicts which CVEs are most likely to be exploited in the future using a machine-learning algorithm that analyzes data from various sources.

Orpheus has trained a Machine Learning algorithm to identify the CVEs (Common Vulnerabilities and Exposures) with the greatest likelihood of being exploited in the future. We collected data sets going back to 2016 to supply a range of features anticipated to be indicative of future exploitation, including the following:

• Dark Web chatter
• Online security researcher communities
• Social media
• Proof of Concept exploits code published in a large range of repositories
• Code to detect whether a system has the vulnerability

Our model was proven to be highly accurate, with 94% accuracy the day before exploitation was confirmed, and 89% three months before.

Download our Ministry of Defence Whitepaper 'Risk Based Vulnerability Management' here to discover how we leverage threat intelligence and machine learning to predict vulnerability exploitation and help organisations to stay protected by doing less patching.

Key Takeaways:

– Our model continuously identifies whether any CVE is one of approximately 0.5% that are actually exploited
– Orpheus’ Machine Learning predicts which CVEs will be exploited in the future with a proven accuracy of up to 94%
– Orpheus’ CVE Risk Score dynamically fuses vulnerability, threat, business impacts and CVE risk, providing your organisation with a concise list of vulnerabilities for remediation