Blind Source Separation in Time and Space
Austrian Science Fund (FWF), grant P31881-N32
There is nowadays an abundance of measurements which are taken at specific locations and / or repeatedly over time. For the resulting multivariate spatial, temporal and spatio-temporal data it can often be assumed that there is a lot of noise in the data and only a few latent signal components carry all the relevant information. Following blind source separation assumptions, the project aims at developing on the one side novel inferential tools to obtain hard formal criteria to decide which components form the signal space, and on the other side support this task by applying visual analytics methods to provide guidance for the iterative refinements and subspace selection. This will be done in an unsupervised and supervised framework and the combination of the statistical methodology together with the visual analytics methods will yield powerful and practical data analysis tools to obtain optimal insights in these complex data structures.
|Nikolaus Piccolotto, Markus Bögl, Christoph Muehlmann, Klaus Nordhausen, Peter Filzmoser, Silvia Miksch, "Visual Parameter Selection for Spatial Blind Source Separation", Computer Graphics Forum, vol. 41, pp. 12, 2022. paper|
|Davide Ceneda, Natalia Andrienko, Gennady Andrienko, Theresia Gschwandtner, Silvia Miksch, Nikolaus Piccolotto, Tobias Schreck, Marc Streit, Josef Suschnigg, Christian Tominski, "Guide Me in Analysis: A Framework for Guidance Designers", Computer Graphics Forum, vol. 39, pp. 19, 2020. paper|
|Markus Bögl, "Visual Analysis of Periodic Time Series Data - Supporting Model Selection, Prediction, Imputation, and Outlier Detection Using Visual Analytics", Institute of Visual Computing and Human-Centered Technology, vol. PhD, Dr.-techn., 2020. paper|