Visual Segmentation and Labeling of Multivariate Time Series

点击翻译按钮获取中文摘要

A highly relevant task in many domains is to find meaningful segments and labels in multivariate time series data to allow humans to generate hypotheses and draw conclusions, for instance to find events or activities in human motion data or electrocardiograph data. Going beyond current segmentation and labeling approaches, this project aims for an interconnected and visual-interactive approach to combine the algorithm selection for segmentation and labeling, the parametrization of these algorithms, and the visualization and exploration of diverse types of uncertainty about and in the results. Current approaches consider each of these problems separately. However, the tight interconnection of these aspects in our combined visual analytics approach will definitely lead to better results as well as to a deeper understanding of the data and the data generating process; in our case with regard to algorithm selection, parametrization, and involved uncertainty in the segmentation and labelling of multivariate time series.A joint system setup, shared sets of data, and task abstractions define a common ground and ensure collaboration throughout the project. We will investigate in the individual aspects with regard to the envisaged interconnections. (a) To open the black-box of algorithm selection, we provide visual analytics techniques to explore the selection of adequate segmentation and labeling algorithms, to steer these algorithms, and to guide the users to detect the most adequate algorithms for a particular data set. (b) We ease the parametrization of these algorithms by developing visual analytics techniques for a systematic analysis of the parameter space (many parameters and large value ranges). (c) For exploring and communicating diverse types of uncertainty, we facilitate appropriate visual encoding, develop visual analytics techniques to assess these types of uncertainty, and allow investigation of uncertainties of alternative algorithms and parametrizations (aggregated uncertainties as well as causes and effects). Such a novel strategy requires a comprehensive evaluation. We plan a horizontal as well as a vertical evaluation strategy. With the horizontal evaluation, we will test single visualization and interaction designs that will be developed during the project. With the vertical evaluation, we will provide a summative evaluation of our combined visual analytics approach.The project team of this German-Austrian collaboration is comprised of experts who have long-standing experience in the three research fields. The University of Rostock (lead Heidrun Schumann) has worked extensively in the field of parametrization; the TU Wien (lead Silvia Miksch) includes experts for time series analysis and uncertainty visualization; and the TU Darmstadt (lead Dieter Fellner) has created several approaches for the data-driven selection of algorithms and the analysis of large sets of time series.