Scientists Use AI to Interlock Computer Vulnerability Databases for Improved Cyber Defense

Researchers from Purdue University, Carnegie Mellon University, Boise State University and the Department of Energy’s Pacific Northwest National Laboratory have interlocked three large databases that contain information on computer vulnerabilities, weaknesses and possible attack patterns to help defenders determine and prevent cyber attacks as quickly as possible. PNNL Chief Computer Scientist Mahantesh Halappanavar said cyber defenders deal with hundreds of information and code that need to be interpreted and prioritized to determine vulnerability and determinative action. The team interconnected an artificial intelligence-based model that automatically links vulnerabilities to specific lines of attacks that could compromise computer systems, PNNL said.

Newswise reported that the model makes use of natural language processing and supervised learning to analyze data from three separate cybersecurity databases to determine potential threats and allow defenders to immediately act on a threat even before it becomes widespread. These databases include the more than 200,000 common vulnerabilities and exposures listed in the national vulnerability database, the 1,000 common weakness enumerations listed in the common weakness enumeration database and the over 500 potential attack routes included in the common attack pattern enumeration and classification resource.