Browsing by Author "Eichner, Christian"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Exploring Time Series Segmentations Using Uncertainty and Focus+Context Techniques
    (The Eurographics Association, 2020) Bors, Christian; Eichner, Christian; Miksch, Silvia; Tominski, Christian; Schumann, Heidrun; Gschwandtner, Theresia; Kerren, Andreas and Garth, Christoph and Marai, G. Elisabeta
    Time series segmentation is employed in various domains and continues to be a relevant topic of research. A segmentation pipeline is composed of different steps involving several parameterizable algorithms. Existing Visual Analytics approaches can help experts determine appropriate parameterizations and corresponding segmentation results for a given dataset. However, the results may also be afflicted with different types of uncertainties. Hence, experts face the additional challenge of understanding the reliability of multiple alternative the segmentation results. So far, the influence of uncertainties in the context of time series segmentation could not be investigated. We present an uncertainty-aware exploration approach for analyzing large sets of multivariate time series segmentations. The approach features an overview of uncertainties and time series segmentations, while detailed exploration is facilitated by (1) a lens-based focus+context technique and (2) uncertainty-based re-arrangement. The suitability of our uncertainty-aware design was evaluated in a quantitative user study, which resulted in interesting findings of general validity.
  • Thumbnail Image
    Item
    Making Parameter Dependencies of Time‐Series Segmentation Visually Understandable
    (© 2020 Eurographics ‐ The European Association for Computer Graphics and John Wiley & Sons Ltd, 2020) Eichner, Christian; Schumann, Heidrun; Tominski, Christian; Benes, Bedrich and Hauser, Helwig
    This work presents an approach to support the visual analysis of parameter dependencies of time‐series segmentation. The goal is to help analysts understand which parameters have high influence and which segmentation properties are highly sensitive to parameter changes. Our approach first derives features from the segmentation output and then calculates correlations between the features and the parameters, more precisely, in parameter subranges to capture global and local dependencies. Dedicated overviews visualize the correlations to help users understand parameter impact and recognize distinct regions of influence in the parameter space. A detailed inspection of the segmentations is supported by means of visually emphasizing parameter ranges and segments participating in a dependency. This involves linking and highlighting, and also a special sorting mechanism that adjusts the visualization dynamically as users interactively explore individual dependencies. The approach is applied in the context of segmenting time series for activity recognition. Informal feedback from a domain expert suggests that our approach is a useful addition to the analyst's toolbox for time‐series segmentation.