Abstract
This paper presents a technique for visualization and interactive level-of-detail control of large number of time series data. The technique first generates clusters of time series values, then selects representative values for each cluster, and finally visualizes only representatives. The technique provides a simplified view without missing its interesting features since it reduces number of displaying polygonal lines by using a clustering algorithm. The technique also provides a user interface so that users can interactively select interesting representatives, and explore the time series values which belong to the clusters of the representatives. We applied the proposed technique to the visualization of air temperature in Japan. This paper introduces that we analyzed weather phenomena by using the proposed level-of-datail control and user interface.
