Alexander Endert
2:00 - 2:30 pm
Assistant Professor in the School of Interactive Computing
Georgia Institute of Technology

Visual Analytics by Demonstration for Helping People Create, Compare, and Use ML Models

Abstract: Machine learning models are powerful and provide impactful insights in a variety of domains. However, to people without formal data science training, knowing how to select the right model, specify the correct parameters, and validate that the model is working are hard tasks. Often, such user interactions are through control panels that provide direct controls over visualization and analytic model parameters. If people do not have expertise in these areas, this interaction design poses fundamental usability challenges.  This talk discusses an alternate interaction methodology, Visual Analytics by Demonstration, that allows people to provide visual demonstrations to interactively explore data in visual analytic applications. I will discuss grounding scientific principles, and give examples of applications to illustrate the idea.

Bio: Alex Endert is an Assistant Professor in the School of Interactive Computing at the Georgia Institute of Technology, where he has been on the faculty since 2014. Endert is a recognized researcher in areas of visual analytics, information visualization, and human-computer interaction, publishing at venues such as IEEE Visual Analytics Science and Technology (VAST), IEEE Information Visualization (InfoVis), ACM Transactions on Computer-Human Interaction (CHI), IEEE Transactions on Computer- Human Interaction (TVCG), and others. His research group (the Visual Analytics Lab) develops ways to empower people with interactive data interfaces to make sense of data. The developed visual analytic systems help people explore, analyze, and discover insights into a broad set of domains. He has served on the Program Committees for IEEE VAST and InfoVis, and on the Organizing Committees for IEEE VIS. In 2015, he was inducted into the Science Advisory Guide for Emergencies for the Department of Homeland Security and Technology Directorate. In 2013, his work on Semantic Interaction was awarded the IEEE VGTC VPG Pioneers Group Doctoral Dissertation Award, and the Virginia Tech Computer Science Best Dissertation Award.

Host: Dana Randall, co-executive director, Institute for Data Engineering and Science; ADVANCE professor of computing, School of Computer Science at Georgia Tech.