Hyperbolic Tree + Time Disc: Visualizing Hierarchical Time-series Data

Abstract

In this paper, we propose a new method of visualizing hierarchical time-series data. We use the hyperbolic tree to visualize the hierarchical structure. The hyperbolic tree can visualize large hierarchical structure. It allocates more space for the nodes of our concern, with the entire hierarchical structure being displayed at the same time. We utilize the time disc, which is similar to the spiral, to display the time-series data. Unlike traditional bar charts and line graphs, the time disc is suited to visualizing large data set and supporting much better the identification of features in the data, such as periodicity and trends. The method can easily display large hierarchical structure and time-series data. We visualize the time series data of each child node by selecting the parent node in the hyperbolic tree, and then observe the similarities and differences between the nodes and the trends of the thing. We applied this method to the urban air quality data visualization and achieved good results.