Caltech Kicks Off Anomaly-Based Data Security Research Program

Faced with a looming tech talent shortage, the technology industry has often worked with academia to engage tomorrow’s professionals, ensure students are learning essential skills and cultivate talent to address emerging cyber threats. In that vein, Guidance Software has funded a program at the California Institute of Technology (Caltech) to support data security research using advanced anomaly science.

Discoveries will be published and used to enhance data breach detection and incident response capabilities.

The joint program will be run out of Caltech’s new lab, the Center for Data-driven Discovery, in collaboration with NASA Jet Propulsion Laboratory (JPL). The Center was established for the advancement of data-intensive, computationally enabled science and technology.

“Hackers are relentless. Breaches of major corporate or government data centers are on the rise to the tune of billions of dollars in losses per year,” said Michael Harris, CMO of Guidance Software. “Our industry is facing a massive labor shortage of cybersecurity specialists. To address this shortfall, we are working with data scientists at Caltech, one of the most respected research universities in the world, to use anomaly detection, complex event processing, and machine learning to help us thwart these breaches and reduce the damage to taxpayers and corporate profits.”

The research will be conducted under the direction of principal scientist Julian Bunn and professor emeritus Mani Chandy, who have developed algorithms and processes to detect anomalous patterns in data sets used in critical areas such as earthquake prediction and now in the detection of advanced malware and their polymorphic variants.

The theory is that timely and effective breach detection of modern threats requires a scientific approach based on machine learning and statistical models. The most common ways to deal with data security issues is through the use of signatures, blacklists or shared threat intelligence. These traditional approaches leave organizations open to the risk of unknown or zero-day malware, which can only be found by focusing on detection of anomalous or atypical behavior that may indicate unauthorized access to sensitive data.

“We are thrilled to be working with Guidance Software on such an important initiative,” said Karina Edmonds, executive director for corporate sponsorships at Caltech. “Our institute is focused on applying science to expand human knowledge and benefit society through research. This work to solve for breach detection and malware discovery is consistent with our mission.”

Source: Caltech Kicks Off Anomaly-Based Data Security Research Program

Via: Google Alert for ML