This project focuses on interactive analysis of complex graphs. Organizations increasingly face a need to understand phenomena from real-world, real-time data sources such as social media, health or financial records, sensor or click streams. Example applications include environmental sensing, patient health and disease monitoring, financial and marketing decisions, etc. Relational data may be represented as time-varying graphs, where nodes represent entities and the edges represent inter-node interactivity or relationships. Both graph structure and attributes of nodes and edges may change as new information arrives or in response to interactive exploration (e.g., filtering or changing edge weight scheme). Visual analysis of the graphs is a challenging problem, especially as they become large. Managers need to integrate multiple graphs, e.g., community networks from social media, consequence analysis networks, and transportation networks. Interactive analytics and visual exploration techniques are key to understanding complex graph structure and behavior. Relevant features need to be extracted and aggregated dynamically based on user input. The primary project goal is to develop integrated high-performance visual analytic techniques that combine graph analysis with visualization and interface techniques for interactive mining of multi-modal, multi-relational graph mining methods.