Visualization allows us to perceive relationships in large data sets. While statistical techniques may determine correlations among the data, visualization helps us frame what questions to ask. A great deal of information visualization tools that attempt to deal with large and complex data produce impressive looking images, and provide powerful interaction possibilities. However, the results are often nearly as complex as the underlying data itself and the learning curve required for productive interaction is steep. And so information visualization tools that are used in practice are traditional ones such as pie-charts, box-plots, and node-and-link graphs. At the visualization group in Arizona we are working on intuitive and easy-to-understand visualization techniques. In this brief talk we'll highlight research by our Humans, Data, Computers (HDC) Lab, led by Katherine Isaacs, Stephen Kobourov, Josh Levine, and Carlos Scheidegger.
