A Visual Analytics Framework for Reviewing Multivariate Time-Series Data with Dimensionality Reduction

Takanori Fujiwara, Shilpika Shilpika, Naohisa Sakamoto, Jorji Nonaka, Keiji Yamamoto, Kwan-Liu Ma

View presentation:2020-10-28T16:00:00ZGMT-0600Change your timezone on the schedule page
2020-10-28T16:00:00Z
Exemplar figure
A screenshot of MulTiDR visual interface. Here we perform an analysis of a dynamic contact network of high school students in Marseilles, France. (a) A two-step dimensionality reduction (DR) view draws the DR results obtained through the two-step DR. (b) A supplemental information view supports understanding selected points in the two-step DR view with the auxiliary information. (c) A feature contribution view visualizes features (either instances, variables, or time points) and their contributions to characteristics of each of the selected clusters. A histogram comparison view shows the feature values in the first DR result of the selected element in (c). (e) A parametric mapping view depicts parametric mappings generated in the first DR, specifically the mappings to the first principal component in this example. (f) The analyst can select a type of DR results.
Keywords

Multivariate time-series, tensor, data cube, dimensionality reduction, interpretability, visual analytics

Abstract

Data-driven problem solving in many real-world applications involves analysis of time-dependent multivariate data, for which dimensionality reduction (DR) methods are often used to uncover the intrinsic structure and features of the data. However, DR is usually applied to a subset of data that is either single-time-point multivariate or univariate time-series, resulting in the need to manually examine and correlate the DR results out of different data subsets. When the number of dimensions is large either in terms of the number of time points or attributes, this manual task becomes too tedious and infeasible. In this paper, we present MulTiDR, a new DR framework that enables processing of time-dependent multivariate data as a whole to provide a comprehensive overview of the data. With the framework, we employ DR in two steps. When treating the instances, time points, and attributes of the data as a 3D array, the first DR step reduces the three axes of the array to two, and the second DR step visualizes the data in a lower-dimensional space. In addition, by coupling with a contrastive learning method and interactive visualizations, our framework enhances analysts' ability to interpret DR results. We demonstrate the effectiveness of our framework with four case studies using real-world datasets.