The national level grand challenge in Computer Science is the ability to acquire, reduce, analyze and extract meaning from massive amounts of data. ORNL is studying methods for the discovery of "hidden laws and processes" underlying observable behaviors in complex systems. ORNL researchers are developing software frameworks that provide a significant advancement in automating the knowledge discovery process. We research and apply methods to efficiently provide comprehensive, accurate, verified, validated, and easily understood information to support situational awareness and informed decision-making, going past traditional approaches to develop new methods for meeting user needs across a wide range of domains. These systems are able to develop actionable insights from massive, dynamic, disparate data while enabling the user to ask more complex questions and detect more complex answers.