VISSECT - Visual Segmentation and Labeling of Multivariate Time Series

Aim

This project's goal is to develop new approaches for visually and interactively analyzing multivariate time series through segmentation and labeling. The ground-breaking novel idea is to combine the algorithm selection, the adequate parametrization, and the visualization and exploration of diverse types of uncertainty about the results.

Team
  • Silvia Miksch
  • Theresia Gschwandtner
  • Markus Bögl
  • Christian Bors
Duration
-
Funding
Contact
Status
finished

The overarching objective of this research project is the combination of the three most relevant steps in the process of segmenting and labeling multivariate time series: algorithm selection, parametrization, and uncertainty analysis. Each of these steps is a research problem on its own, and we will contribute novel solutions to all of them. However, only the combination of the different aspects will comprehensively address the important challenge of visual analytics as identified by Jarke van Wijk: "Which model? Which parametrization? And which feature of the data?" In the following we will describe the underlying research questions for the individual aspects with regard to the envisaged interconnections.

Algorithm selection: Segmentation and labeling algorithms divide multivariate time series into smaller segments and label these segments accordingly. The effect of a particular algorithm on a particular data set is not easily predictable and thus finding the best algorithm in a great diversity of existing algorithms is highly demanding.

Parametrization: The effect of different parametrizations on the segmentation process is not easy to understand, and thus finding the best setting is challenging. Particularly in the case of large parameter spaces (many parameters and large value ranges) one typically has to rely on trial and error to find adequate configurations. So, the main goal of this objective is to develop sophisticated visual analytics techniques for a systematic analysis of the parameter spaces.

Uncertainty: The generated segmentation and labeling results may comprise different kinds of uncertainty at different levels. These stem from the selection of algorithms, parameters, and the calculation of multiple competing results.

In sum, our major objective is to investigate each single aspect (algorithms selection, parametriza- tion, and uncertainty assessment) with regard to the other aspects, to observe the interplay, to combine them, and by doing so, to develop a ground-breaking novel visual analytics approach for analyzing time series data.
 
The three main challenges in the process of segmenting and labeling multivariate time series data.

Publications

Christian Bors, Christian Eichner, Silvia Miksch, Christian Tominski, Heidrun Schumann, Theresia Gschwandtner, "Exploring Time Series Segmentations Using Uncertainty and Focus+Context Techniques", EuroVis 2020, 2020. paper
Christian Bors, "Facilitating Data Quality Assessment Utilizing Visual Analytics: Tackling Time, Metrics, Uncertainty, and Provenance", Institute of Visual Computing and Human-Centered Technology, vol. PhD, Dr.-techn., 2020.
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
An uncertainty quantification cube for multivariate time series Christian Bors, Jürgen Bernard, Markus Bögl, Theresia Gschwandtner, Jörn Kohlhammer, Silvia Miksch, "Quantifying Uncertainty in Multivariate Time Series Pre-Processing", EuroVis Workshop on Visual Analytics (EuroVA), 2019.
Jürgen Bernard, Marco Hutter, Heiko Reinemuth, Hendrik Pfeifer, Christian Bors, Jörn Kohlhammer, "Visual-Interactive Preprocessing of Multivariate Time Series Data", Computer Graphics Forum, vol. 38, pp. 11, 2019. paper
The Quality Flow and Provenance Graph views of the Data Quality Provenance Explorer. Christian Bors, Theresia Gschwandtner, Silvia Miksch, "Capturing and Visualizing Provenance From Data Wrangling", IEEE Computer Graphics and Applications, vol. 39, pp. 15, 2019. paper
Iterative refinement of the provenance task abstraction framework. Christian Bors, John Wenskovitch, Michelle Dowling, Simon Attfield, Leilani Battle, Alex Endert, Olga Kulyk, Robert Laramee, "A Provenance Task Abstraction Framework", IEEE Computer Graphics and Applications, vol. 39, pp. 15, 2019. paper
Fabian Schwarzinger, Andreas Roschal, Theresia Gschwandtner, "Sketching Temporal Uncertainty - An Exploratory User Study", EuroVis 2018 - Short Papers, Eurographics/IEEE VGTC Conference on Visualization, pp. 67-71, 2018. paper supplemental material
Jürgen Bernard, Christian Bors, Markus Bögl, Christian Eichner, Theresia Gschwandtner, Silvia Miksch, Heidrun Schumann, Jörn Kohlhammer, "Combining the Automated Segmentation and Visual Analysis of Multivariate Time Series", EuroVis Workshop on Visual Analytics (EuroVA) 2018, pp. 49–53, 2018. paper
Christian Bors, Theresia Gschwandtner, Silvia Miksch, "Visually Exploring Data Provenance and Quality of Open Data", EuroVis 2018 - Posters, pp. 9–11, 2018. paper
Markus Bögl, Christian Bors, Theresia Gschwandtner, Silvia Miksch, "Categorizing Uncertainties in the Process of Segmenting and Labeling Time Series Data", , 2018. paper
Markus Bögl, Christian Bors, Theresia Gschwandtner, Silvia Miksch, "Uncertainty types in segmenting and labeling time series data", , 2018. paper
Christian Bors, Markus Bögl, Jürgen Bernard, Theresia Gschwandtner, Silvia Miksch, "Quantifying Uncertainty in Time Series Data Processing", , 2018.
Markus Bögl, Peter Filzmoser, Theresia Gschwandtner, Tim Lammarsch, Roger Leite, Silvia Miksch, Alexander Rind, "Cycle Plot Revisited: Multivariate Outlier Detection Using a Distance-Based Abstraction", Computer Graphics Forum, vol. 36, pp. 227–238, 2017. paper Usage Scenario
Christian Bors, Markus Bögl, Theresia Gschwandtner, Silvia Miksch, "Visual Support for Rastering of Unequally Spaced Time Series", 10th International Symposium on Visual Information Communication and Interaction (VINCI), pp. 53-57, 2017. paper
Christian Bors, Markus Bögl, Theresia Gschwandtner, Silvia Miksch, "Visual Support for Rastering of Unequally Spaced Time Series", Data Science, Statistics & Visualisation Conference (DSSV), 2017. paper
Jürgen Bernard, Eduard Dobermann, Markus Bögl, Martin Röhlig, Anna Vögele, Jörn Kohlhammer, "Visual-Interactive Segmentation of Multivariate Time Series", EuroVis Workshop on Visual Analytics (EuroVA), 2016. paper
Theresia Gschwandtner, Heidrun Schumann, Jürgen Bernard, Thorsten May, Markus Bögl, Silvia Miksch, Jörn Kohlhammer, Martin Röhlig, Bilal Alsallakh, "Enhancing Time Series Segmentation and Labeling Through the Knowledge Generation Model", Poster Proceedings of the Eurographics Conference on Visualization (EuroVis 2015), pp. 3, 2015.