DataBurst: Interactive Analysis of Hierarchical Data Using Radial Space-Filling Diagrams

Radial visualization methods have been used frequently in the field of information visualization due to their compact design and aesthetic layout of data items that can often aid in the communication of relationships between them. Published methods offer varying levels of effectiveness however, due to restrictions in design, data and interactivity. Their use is often restricted to particular data types, perspectives and interfaces that may not completely match the needs of the analysis, facilitate detailed data investigation or allow for knowledge discovery outside of what was originally intended when the visualization was developed. We therefore present DataBurst, an open, flexible data visualization that extends traditional concentric radial space-filling diagrams using two methods of Focus+Context (Hyperbolic Distortion and TearAway Subdiagrams) and uniquely leveraging three radial diagram characteristics. DataBurst's effectiveness with varying volumes and types of data is then shown with case studies in two active areas of research (network security and bioinformatics) and show that insight can be quickly gained by interactively viewing up to three attributes of data at once in the context of the hierarchy of the data.