Visual Segmentation and Labeling of Multivariate Time Series

在许多领域中，一个高度相关的任务是在多变量时间序列数据中找到有意义的片段和标签，以允许人类生成假设并得出结论，例如在人体运动数据或心电图数据中找到事件或活动。超越目前的分割和标记方法，这个项目的目标是一个相互关联的和视觉交互的方法，结合联合收割机的分割和标记，这些算法的参数化，并可视化和探索各种类型的不确定性和结果。目前的方法分别考虑这些问题中的每一个。然而，在我们的组合可视化分析方法中，这些方面的紧密互连肯定会带来更好的结果，以及对数据和数据生成过程的更深入理解;在我们的情况下，关于算法选择，参数化，以及多变量时间序列的分割和标记中的不确定性，联合系统设置，共享数据集，和任务抽象定义了一个共同点，并确保整个项目的协作。我们将在各个方面进行调查，以了解所设想的相互联系。(a)为了打开算法选择的黑盒子，我们提供了可视化分析技术来探索适当的分割和标记算法的选择，引导这些算法，并指导用户为特定数据集检测最合适的算法。(b)我们通过开发可视化分析技术来系统地分析参数空间（许多参数和大值范围），从而简化了这些算法的参数化。(c)为了探索和交流各种类型的不确定性，我们促进适当的视觉编码，开发视觉分析技术来评估这些类型的不确定性，并允许调查替代算法和参数化的不确定性（汇总的不确定性以及原因和影响）。这种新的战略需要进行全面的评价。我们计划一个横向和纵向的评估策略。通过横向评估，我们将测试项目期间开发的单个可视化和交互设计。通过垂直评估，我们将对我们的综合视觉分析方法进行总结性评估。此次德奥合作的项目团队由在这三个研究领域拥有长期经验的专家组成。罗斯托克大学（Heidrun Schumann领导）在参数化领域开展了广泛的工作; TU Wien（Silvia Miksch领导）包括时间序列分析和不确定性可视化方面的专家; TU达姆施塔特（Dieter Fellner领导）为数据驱动的算法选择和大型时间序列集的分析创建了几种方法。

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.