December 12, 2017 • Chris Pace
Scaling human analysts to process the sheer volume of available threat data is impossible. That’s why we built the Threat Intelligence Machine, combining the power of cognitive systems with experienced and expert analysts to deliver threat intelligence insights.
This short series of videos explains how we apply machine learning to the problem of gathering, analyzing, and structuring threat intelligence in real time.
Recorded Future experts explain what machine learning is and how we apply it to the organization and the analysis of unstructured data.
The Threat Graph describes how we use machine learning to represent the world of cyber threats and attacks. The Threat Intelligence Machine creates relationships in the data from a huge breadth of varied threat data sources, powering faster analysis and more informed security decisions.
We’re training machines to read and understand words in threat data in multiple languages and at unparalleled scale. Our machine automatically highlights attack methods, targets, vulnerabilities, and exploits, cutting through the noise to deliver relevant threat intelligence.
New threats are uncovered each day, but confusing and duplicated naming makes it hard to understand available intelligence. We’ve trained a machine to predict where text appearing in threat data refers to new malware names, as well as deduplicate where aliases are invented.
We’re applying machine learning to generate predictive models that can be used to forecast events or classify entities. For example, we have created models to predict the likelihood of product vulnerabilities being exploited, and to assess the risk that an IP address will behave maliciously in the future, even if no such activity has been seen yet.
You can get more detailed information on how these machine learning techniques power a comprehensive view of the whole threat landscape in our new white paper, “4 Ways Machine Learning Is Powering Smarter Threat Intelligence.”