NEW YORK, Aug. 31, 2016 /PRNewswire/ -- SecurityScorecard, the most accurate Security rating and continuous risk monitoring platform, today announced a year-long partnership with Columbia University's Data Science Institute. The two will collaborate on various data science and machine learning projects to build breach prediction models that will help determine how and why cyberattacks happen, and the types of organizations that are vulnerable to them.
Breach prediction models provide a way to assess the Security risk for clients as well as other companies. The objective of the analysis is to investigate methods to develop learning algorithms to automatically cluster organizations into groups based on the malware and port vulnerability features in their data sets – vital proprietary information SecurityScorecard uses to understand how cyberattacks happen.
Scores created by SecurityScorecard already cover the security hygiene of a large number of companies, based on a wide breadth of security indicators. By identifying which types of security vulnerabilities are more closely tied to actual breaches or attacks, SecurityScorecard aims to further improve its scoring, which is already top in the field.
"SecurityScorecard is constantly looking for ways to improve the predictive power and accuracy of our scores, and this requires a variety of approaches. By partnering with Columbia, we will expand the resources available for research using advanced data science techniques," said Luis Vargas, Senior Data Scientist at SecurityScorecard. "Research outcomes from the partnership will be used to improve our analytics, and gain insights into new ways of developing algorithms."
The university plans to use the data as a starting point to build risk models for cyber insurance.
"This is an exciting partnership between the university and SecurityScorecard. Columbia will base its analysis on SecurityScorecard's rapidly growing IP-based threat and vulnerability data such as insecure ports and malware infections, in significantly more detail than has been done before," said Vishal Misra, Computer Science Professor at Columbia Engineering. "Therefore, we expect that new insights will be uncovered from this data. We cannot wait to get started."
SecurityScorecard chose to partner with Columbia University given their wealth of expertise in machine learning, as well as Security.
"We believe that Columbia has the right combination of expertise, mentorship, and committed students who can apply their skills to conduct research on our data," said Vargas. "As a leading startup in New York City, SecurityScorecard also has a strong interest in helping create connections with the next generation of data scientists here in the city."
SecurityScorecard provides the most accurate rating of security risk for any organization worldwide. The proprietary SaaS platform helps enterprises gain operational command of the security posture for themselves and across all of their partners, and vendors. It provides continuous, non-intrusive monitoring for any organization including third and fourth parties. The platform offers a breadth and depth of critical data points not available from any other service provider including a broad range of risk categories such as Application Security, Malware, Patching Cadence, Network Security, Hacker Chatter, Social Engineering and Passwords Exposed.