Michael Gleicher
University of Wisconsin-Madison, WI·Computer Science
When a dataset holds thousands of complex items, such as documents, medical records or even chairs, averages and standard charts often say little that a person can grasp. One simple alternative is to show a handful of well-chosen examples.
- Fields
Computer Science
- Topics
data visualization
information visualization
subset selection
- Grants
- 1 known grant, last ended Sep 2026
What they work on
- Determine how people interpret small subsets of data items to grasp an overview of a large collection, using empirical user studies that link subset design choices to task performance.
- Develop algorithms that embed complex objects, such as documents or records, into metric spaces and sample representative subsets from them for summary visualization.
- Create table-based displays and interaction techniques that let viewers browse a large dataset by changing which subset of items is shown.
- Package subset summarization methods into open-source visualization toolkits and write guidelines advising practitioners when and how to use subset selection.
More about the research
This is called a subset summary. Yet little is known about how people read such subsets to get an overview. There are also no guidelines for choosing which items to show or how to display them. This project starts from the premise that understanding how viewers interpret subsets will let visualization designers build more effective summaries.
Why it matters
Many fields need to make sense of large collections of complex items, such as documents or medical records, that standard statistics summarize poorly. Evidence-based methods and open-source tools for choosing and displaying representative examples could help analysts and visualization designers across these domains. These tools could help them build clearer overviews of their data.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Christopher Bryan
Arizona State University Campus Immersion, AZ·Computer Science
Develop a visualization grammar, a way to specify chart designs without low-level programming, that supports privacy constraints and data sanitization actions.
Alexander Kale
University of Chicago, IL·Computer Science·Early-career
Develop a theory that defines visualization design goals as trade-offs among kinds of information loss, made computable through mathematical formalisms derived from example analysis, synthesis and expert interviews.
Haeyong Chung
University of Alabama in Huntsville, AL·Computer Science·Early-career
Analyze existing multi-display applications in the academic literature to derive design knowledge on how multiple displays and devices can form one coherent visualization system for visual analysis.
The school
University of Wisconsin-Madison: 91 labs on PI Pages, $1.9B a year on research, 854 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.