We are investigating a fast layout method for visualizing and exploring relationships between multivariate data items. We improve on existing works that use the force-directed layout, which has high running time and cannot scale up for large-scale visual analysis. Our method, based on Mean Value Coordinates, has a closed-form solution that can determine items' locations in a single iteration. In addition, it has a fast running time that is linear in the number of items. We are also exploring

more »

... tiple interactive visualization techniques to help users make sense of the data, such as blending multiple heat maps to simultaneously express multiple types of data distributions; and techniques to create topics, and to merge or split topics in real time.