The paper elaborates on the previous research on the analysis of temporal and spatio-temporal data done in statistical graphics and geo-visualization. We focus on the exploration of spatially distributed time-series data, i.e. values of numeric attributes referring to different moments in time and locations in space. After considering appropriate interactive visualization techniques, we propose several methods of exploration based on user-controlled data transformation and aggregation, easy-to-understand calculations, and dynamic linking of data displays. The proposed methods are potentially scalable, i.e. they can be applied to large data sets without overcrowding the displays and loosing interactivity.
Gennady L. Andrienko, Natalia V. Andrienko