Universities To Help Train Watson on Cybersecurity

• By Dian Schaffhauser
• 05/10/16

Eight North American universities will be working with IBM to train Watson on cybersecurity. The artificial intelligence system has already begun "learning" security research related to spam and phishing attacks and documented vulnerabilities compiled over the last two decades by the company. Students at the institutions will help train Watson on the "language" of cybersecurity by annotating security reports and data and loading those into the system.

Among the universities participating are California State Polytechnic University, Pomona; the University of Waterloo; and the University of Maryland, Baltimore County (UMBC).

This training effort, which will give the participants practical experience in a field known as "cognitive security," is intended to help build up Watson's cybersecurity taxonomy and its understanding of infection methods, indicators of compromise and signs of potential advanced persistent threats.

Later this year IBM expects to be able to run beta deployments of Watson for customer problems that tap the system's analytics functionality on cybersecurity problems. Eventually, Watson for Cybersecurity, as it's being called, will use natural language processing to analyze and uncover patterns in both structured and unstructured data, such as security research papers, blogs, articles, videos and alerts. The goal: to generate insights into emerging threats and provide recommendations on how to stop them. An example offered by IBM is the use of Watson to tie data on an emerging form of malware in an online security bulletin with other data from a security analyst's blog on an emerging remediation strategy.

"Providing our students an opportunity to work one-on-one with Watson will undoubtedly set them apart when it's time to enter the workforce," said Dan Manson, chair of the computer information systems department at Cal Poly Pomona, in a prepared statement. "Not only is there potential to search for and repair vulnerabilities, Watson's ability to recognize advanced patterns could fill holes we didn't even know existed."

Related to the initiative, UMBC will work with IBM Research to create a new lab this fall in its college of engineering and IT. The Accelerated Cognitive Cybersecurity Laboratory will enable faculty and students to work with IBM scientists on the company's advanced computing systems to address cybersecurity research problems.

The new lab will be led by Anupam Joshi, who heads up the university's Center for Cybersecurity and chairs the department of computer science and electrical engineering. His multidisciplinary team of faculty members, graduate and undergraduate students and software engineers will bring together expertise in the fields of cognitive computing, accelerated and high performance computing, and cybersecurity.

They'll be working on IBM Power Systems servers using OpenPower technology with support from IBM's Systems Group.

"The volume and velocity of data in security is one of our greatest challenges in dealing with cybercrime," noted Marc van Zadelhoff, general manager for IBM Security. "By leveraging Watson's ability to bring context to staggering amounts of unstructured data, impossible for people alone to process, we will bring new insights, recommendations and knowledge to security professionals, bringing greater speed and precision to the most advanced cybersecurity analysts, and providing novice analysts with on-the-job training."