Visual‐interactive Exploration of Interesting Multivariate Relations in Mixed Research Data Sets

Abstract

The analysis of research data plays a key role in data‐driven areas of science. Varieties of mixed research data sets exist and scientists aim to derive or validate hypotheses to find undiscovered knowledge. Many analysis techniques identify relations of an entire dataset only. This may level the characteristic behavior of different subgroups in the data. Like automatic subspace clustering, we aim at identifying interesting subgroups and attribute sets. We present a visual‐interactive system that supports scientists to explore interesting relations between aggregated bins of multivariate attributes in mixed data sets. The abstraction of data to bins enables the application of statistical dependency tests as the measure of interestingness. An overview matrix view shows all attributes, ranked with respect to the interestingness of bins. Complementary, a node‐link view reveals multivariate bin relations by positioning dependent bins close to each other. The system supports information drill‐down based on both expert knowledge and algorithmic support. Finally, visual‐interactive subset clustering assigns multivariate bin relations to groups. A list‐based cluster result representation enables the scientist to communicate multivariate findings at a glance. We demonstrate the applicability of the system with two case studies from the earth observation domain and the prostate cancer research domain. In both cases, the system enabled us to identify the most interesting multivariate bin relations, to validate already published results, and, moreover, to discover unexpected relations.

A Review of Guidance Approaches in Visual Data Analysis: A Multifocal Perspective

“…Scorpion [WM13] is a tool that supports the exploration of data outliers, by pointing users to the possible data tuples from which these outliers originated. Finally, Bernard et al [BSW∗14] developed a system that emphasizes the most interesting relations among data subsets, thus helping the user gaining an overview of the dataset. Gladish et al [GST13] developed an approach that, by using a flexible degree‐of‐interest measure, is able to show interesting data regions to explore during the analysis of hierarchical data (see Figure 6).…”

Using Dashboard Networks to Visualize Multiple Patient Histories: A Design Study on Post-Operative Prostate Cancer

“…The criterion for Figure 1 is that there is a single bin for all segments, whereas in Figures 3 and 4 the aggregation routine creates a bin for every treatment combination. In general, if the criterion is based on a categorical attribute, we recommend that categorical attributes are binned so that the most frequent categories are assigned to individual bins, and remaining categories are grouped together [43]. For numerical attributes (e.g., PSA), the bins are defined according to standard binning variants which can be domain-preserving (the default in the present paper), frequencypreserving, or be based on a goodness-of-fit measure [44], [45].…”

FuryExplorer: visual-interactive exploration of horse motion capture data

“…In particular, a selection-change event triggers the Sequence Detail List View to refresh the list of visible elements R 9 . In order to explore hidden relations between the data content (poses/gaits) and the metadata (like horses' names, health/lameness status, or variation of hoof motion), 39,40 the Meta Data Viewer lists available metadata attributes, as can be seen on the right-hand side of Figure 4. In this example, variation within the movement of the feet of four different horses is illustrated by colored bar charts.…”