Visual Assistance for Importing Time-oriented Data Tables
Problem
Visual analytics of time-oriented data are common and important for numerous application scenarios, e.g., in healthcare, business, and environmental studies. Furthermore, time and time-oriented data have distinct characteristics that make it worthwhile to treat it as a separate data type. Before any Visual Analytics processes can be performed, however, these data need to be imported. This will be particularly tricky, if the data are available but not their characteristics, the metadata. Therefore an interactive visual interface is needed to specify such metadata based on an overview of the data.
Examples:
- Column 1 e.g., "03.05.2014” + Column 2 e.g., "18:10” → time point of temporal granularity minutes
- Column 1 e.g., "15.07.2014” + Column 2 e.g., "02.08.2014” → time interval of temporal granularity days
- row number → time point of temporal granularity year starting with 2001 (based on background knowledge of the dataset)
Aim
Design and prototypical implementation of an interactive visual interface for importing time-oriented data from tables (e.g., CSV files) based on the software library TimeBench. The project scope includes architectural provisions for future extensions.
Other information
Resources:
Introductory reading:
Aigner, W., Miksch, S., Schumann, H., & Tominski, C. (2011). Visualization of Time-Oriented Data. Berlin: Springer. (esp. chapter 3)
Rind, A., Lammarsch, T., Aigner, W., Alsallakh, B., & Miksch, S. (2013). TimeBench: A Data Model and Software Library for Visual Analytics of Time-Oriented Data. IEEE Transactions on Visualization and Computer Graphics, 19(12), 2247–2256. doi:10.1109/TVCG.2013.206
Nocke, T. (2000). Metadatengewinnung und -spezifikation für Visualisierungsentscheidungen. Diploma thesis, University of Rostock, Germany.