User Behavior Analytics at Netflix – A Data Driven Approach to Identify Information Security Risk

Abstract: User behavior analytics (UBA) is an emerging set of security controls and processes that inspect patterns of user behavior and apply various methods to identify anomalies and events within that behavior representative of security threats. Rather than focus on networks, systems, or applications, UBA centers on the user, particularly user access of cloud data, and is intended to identify malicious insiders, compromised users, fraud, and various advanced attacks.

In this session, we’ll talk about the Netflix solution for UBA, including the data processing architecture, various analytic components that identify abnormal patterns, and systems that drive appropriate action. In addition, we will share some critical success factors and learnings as we embarked on this journey.

Bio: Rekha is a Security Engineer at Netflix with expertise in security, machine learning and distributed systems. Her forte is combining system analysis with machine learning to develop innovate security solutions. She has a PhD in Computer Science from Rutgers University. Before joining Netflix, she was a Research Scientist at Intel Labs working on Security Analytics and Adversarial Machine learning.