Applying Machine Learning to Historical Cyber Attack Data to Predict and Prevent Future Attacks

This research demonstrates that machine learning algorithms can leverage historical data to predict the nature and severity of future cyber attacks with reasonable accuracy, revealing that despite the sophistication of state-sponsored actors, their attack patterns often exhibit digital fingerprints that offer valuable clues for anticipating future threats amidst global geopolitical tensions.

Geopolitical tensions are high, with instability and uncertainty in multiple conflict zones which ripple around the world to countries that support the rule of law, those that do not, and those that are otherwise affected by the sprawling crises. While these conflicts are catastrophic, they are also convenient cover for state-sponsored actors to launch cyber attacks to damage critical infrastructure, undermine democratic institutions and values, and gather sensitive information to prepare for future attacks. State-sponsored actors have become highly sophisticated, with capabilities that leverage the power of technology to prey on governments and companies. Yet despite this sophistication, their attack patterns often have similar digital fingerprints. This reality creates the possibility that past attacks may provide clues for future ones. This paper, which is based on a machine learning algorithm, finds that historical data alone can be sufficient to predict the nature of future attacks with reasonably reliable accuracy. In particular, the severity of the previous attacks gave reliable insight into the severity of future attacks.

Explore More!

Research Paper