Visual Pattern Exploration At and Across Scales
Fritz Lekschas (Harvard University)
Colloquium
Monday, February 22, 2021, 3:30 pm
Abstract
Visually exploring data is a powerful approach to discover, understand, and interpret novel or not-well defined patterns. It allows us to gain insights and generate hypotheses for subsequent analyses. However, visual exploration can become challenging when the patterns of interest are sparsely-distributed, several orders of magnitude smaller than the entire dataset, or detected with high uncertainty.
In this talk, I will present new visualization systems and interaction techniques for efficiently browsing, comparing, and finding patterns in the context of genomic, geospatial, and time-series data. Specifically, I will describe a web platform for browsing multi-modal and multi-scale datasets, as well as their guided navigation. I will then present a generalized framework and toolkit for interactively arranging, grouping, and aggregating thousands of pattern instances. Further, I will demonstrate how interactive visual machine learning can enhance our ability to find patterns effectively. In combining visualization and human-centered machine-learning, these systems ensure that human-in-the-loop data analysis remains feasible with increasingly-large and complex scientific datasets.
Bio
Fritz Lekschas is a Ph.D. candidate in computer science at Harvard University, where he is advised by Hanspeter Pfister. His research focuses on the development of scalable visual exploration systems for analyzing patterns in biomedical data. Prior to his doctoral program, Fritz visited Harvard Medical School as a post-graduate research fellow to work with Nils Gehlenborg and Peter J. Park on ontology-guided exploration of biological data repositories. He earned his bachelor’s and master of science degrees in bioinformatics from the Freie Universität Berlin, Germany. Fritz's work has been recognized with several awards, including a Siebel Scholars Award, the Best Paper Award from EuroVis 2020, and a Best Paper Honorable Mention from IEEE InfoVis 2020.