Knowledge workers deal with incomplete, disparate, conflicting, and uncertain data to address challenging problems that cannot be addressed easily through automated approaches. Human insight is needed to address these problems. Whether engaged in discovery, situation awareness, or decision-making, knowledge workers must sift through overwhelming amounts of data. Their objective may be to understand a changing situation, spot trends and relationships, detect anomalies, discover new insights, make reliable recommendations or decisions. Visual analytics combines innovative interactive visualizations with advanced automated data analysis techniques to enable users to gain deeper insights from their data. PNNL has pioneered visual analytics and is driving toward next-generation capabilities that create powerful partnerships between knowledge workers and their systems.
CAPABILITIES

Graph and Network
Many complex science and security problems can be described in terms of graphs and networks, using nodes and links to represent objects and their relationships. Mining insight from these large, complex graphs requires interactive visualizations that present the graph features and results of automated analytics in ways that can be interpreted straightforwardly by users. Our approach combines intuitive approaches for visual exploration, analysis, and discovery that can be applied to very large graphs in disciplines ranging from security and critical infrastructure protection to bioinformatics and Earth science.

Immersive Computing
This multidisciplinary capability explores future interaction and visualization techniques for data analytics. Recent advances in immersive technologies, including virtual and augmented reality, provide added opportunities to explore, discover, and analyze data that augment current analytic methods and decision support. Deriving insights from data by enhancing the user experience with new display, input, and interaction techniques can increase user engagement and productivity. This research examines the application and development of emerging user-interface technologies for creating more engaging experiences and seamless workflows for data analysis applications.

Human Machine Teaming
Remaining at the forefront of developing the next-generation visual analytic systems requires advances in machine learning and artificial intelligence combined with new interactive user interfaces that will enable users to gain insight into increasingly complex problems. These new systems will represent a paradigm shift: instead of forcing users to express their questions as a series of queries, the system supports the user, inferring interests from interactions. The result is that users can focus on tasks and think about problems rather than continuously translating between analytic questions and data.

Situation Awareness
Protecting security and critical infrastructure is a 24/7 job, and operations centers must maintain situation awareness to identify issues as they arise. PNNL’s situation awareness visualizations display meaningful patterns from real-time operations center data in ways that are straightforward to interpret, providing rapid insight into emerging conditions. PNNL software is used in both government and industry environments.

Text and Media
PNNL has amassed more than 20 years of experience in developing first-of-a-kind tools that support interactive exploration and discovery in diverse data collections, including unstructured text, structured data, image collections, and video. We combine innovations in feature extraction, intuitive visualizations, and interactive support for critical thinking. The resulting tools are used by analysts in both government and industry to assess complex data to make effective decisions.