Researchers will have access to ‘real-world’ data on cyber-attacks across the Internet

The Science and Technology Directorate at DHS is planning to gather “comprehensive real-world data” about cyber-attack phenomena, so it can make those data sets available to an estimated 200 different research organizations attempting to develop products and technologies that can better protect the nation’s computing infrastructure.

Under a program called Protected Repository for the Defense of Infrastructure Against Cyber Threats, or PREDICT, the S&T directorate seeks to make these real-world data sets widely available to researchers, who today have to rely instead on anecdotal or small-scale test experiments.

“The data sets are intended to provide developers with timely and detailed insight into cyberattack phenomena occurring across the Internet and in some cases will reveal the effects of these attacks on networks that are owned or managed by the data producers,” says a notice published by S&T in the Federal Register on September 1.

Research groups will be asked to apply for access to the PREDICT data sets by completing a newly-assembled package of forms prepared by S&T. “In addition to helping to determine whether a group is eligible to access the repository, the forms will also manage the interactions between the PREDICT portal administrators and the research groups accessing the PREDICT portal,” says the S&T notice.

The Web portal’s URL is https://www.predict.org. The Coordinating Center for PREDICT will manage the centralized repository, and act as gatekeeper for access to and release of the data. “All data input to the system is either keyed in by users (Data Providers) or migrated (via upload of XML files),” says the notice.

DHS has estimated that 206 research organizations will complete the required package of forms, and that it will take each organization an average of eight hours to do so.

Further information is available from Jeffery Harris at 202-254-6015.

“Researchers, software developers, inventors associated with an authorized sponsoring institution may be granted access to the PREDICT portal,” explains the PREDICT Web site. “In addition, organizations that provide data to the researchers or host the data may hold accounts.

“PREDICT datasets are available to approved Researchers who are conducting cyber security research that is in the interests of the United States,” the Web site continues. “All research and work involving PREDICT datasets must be carried out at locations within the 50 United States.”