“VA, ML, and Workflow Optimization”

Min Chen

Professor of Scientific Visualization at Oxford University

Abstract: Visual analytics (VA) is an integral part of data science. During the first decade of VA, VA researchers focused on the integration of interactive visualization with data mining algorithms, many of which were machine-learned (ML) models. In the past (second) decade, more effort was made to support the development of ML models, e.g., for understanding, explaining, and improving models. When we consider ML models only, they were mainly USED in the VA workflows in the first VA decade, whilst they were also DEVELOPED in the second VA decade. In this talk, the speaker will provide an analysis of these two types of VA workflows with the aid of theoretical abstraction and practical examples, and will demonstrate that VIS4ML (or VA4ML) can benefit the development of ML models hugely as well as being an exciting playground for VA research. The new challenges are not just about technical integration, but more about workflow optimization.

Speaker Biography: Min Chen developed his academic career in Wales between 1984 and 2011. He is currently Professor of Scientific Visualization at Oxford University and a fellow of Pembroke College. His research interests include many aspects of data science in general, and visualization and visual analytics (VIS) in particular. His recent contributions in VIS include topics such as theory of visualization, visual analytics for machine learning, and perception and cognition in visualization. He has worked on a broad spectrum of interdisciplinary research topics, ranging from the sciences to sports, and from digital humanities to cybersecurity.