Smart Explore

Smart Explore


Simplifying High-Dimensional Data Analysis
through a Table-Based Visual Analytics Approach

Simplifying High-Dimensional Data Analysis through a Table-Based Visual Analytics Approach



Abstract

We present SMARTexplore, a novel visual analytics technique to simplify the identification and understanding of clusters, correlations, and complex patterns in high-dimensional data. The analysis is integrated into an interactive table-based visualization, keeping a consistent and familiar representation during the analysis. The visualization is tightly coupled with pattern detection, subspace analysis, reordering, and layout algorithms. To increase the analyst's trust in the revealed patterns, SMARTexplore automatically selects and computes statistical measures based on dimension and data properties. While existing approaches to analyze high-dimensional data (e.g., planar projections and parallel coordinates) have proven to be effective, they typically have steep learning curves for non-visualization experts. Our evaluation, based on expert case studies, confirms that non-visualization experts successfully reveal patterns in high-dimensional data when using the SMARTexplore technique.

...
...

Supplementary Material

Paper, Videos,
and Presentation

Running Prototype
and Source Code

Datasets

Team

Meet the team working for Smart Explore

...
Michael Blumenschein
University of Konstanz, Germany

...
Michael Behrisch
University of Harvard, US

...
Stefanie Schmid
University of Konstanz, Germany

...
Simon Butscher
University of Konstanz, Germany

...
Deborah Wahl
University of Konstanz, Germany

...
Karoline Villinger
University of Konstanz, Germany

...
Britta Renner
University of Konstanz, Germany

...
Harald Reiterer
University of Konstanz, Germany

...
Daniel Keim
University of Konstanz, Germany

Funding Information

We wish to thank the German Research Foundation (DFG) for their financial support within the projects A03 & C01 of SFB/Transregio 161. Furthermore, this research was supported by the Federal Ministry of Education and Research within the research project SmartAct (BMBF Grant 01EL1820A) .