Making Parameter Dependencies of Time‐Series Segmentation Visually Understandable

Abstract

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.

Visual Neural Decomposition to Explain Multivariate Data Sets

“…For geospatial data sets, specific methods were developed that facilitate maps [19,35]. Several approaches rank variables and pairs of variables according to statistical correlation factors [15,35,44] or classification metrics such as separability [52]. Barlowe et al [5] visualize partial derivatives of the dependent variable with regard to the independent variables and analysts can iteratively explore multi-correlations.…”

A Survey of Visual Analytics Techniques for Machine Learning

“…Before Model Building Improving Data Quality (31) [3], [11], [14], [16], [17], [18], [25], [45], [61], [91], [96], [101], [102], [118], [123], [125], [136], [144], [157], [193], [202], [204], [205], [214], [228], [229], [232], [257], [259], [268], [275] Improving Feature Quality (6) [109], [132], [184], [195], [223], [239] During Model Building Model Understanding (30) [28], [38], [56], [71], [79], [84], [104], [115], [116], [119], [120], [137],…”

Multi-display Visual Analysis: Model, Interface, and Layout Computation

“…The analysis addresses two objectives: finding out what properties of the segmentation are influenced and determining parameters that are exerting influence [Eichner et al, 2019]. Here, we describe a parameter-first analysis strategy, for which the starting point are the parameters.…”

“…In addition, a dedicated view shows the variation of parameter influences. This view is based on calculated correlations, more precisely on deviations of correlations computed for parameters and features of the segmented time series [Eichner et al, 2019].…”

“…First, the overviews of high-influence parameters are placed on the same display, and the overviews sorted according to parameters with little influence are removed so that more display space is available (see Figure 9). Additionally, for each important parameter, a triangular subrange visualization of correlation values is called up as a new view to help analysts assess the influence in different parameter sub-ranges [Eichner et al, 2019]. Finally, a parallel coordinate plot is added to support the comparison of the average correlation strength of the three remaining parameters for segments with different labels.…”