Visual Analytics for High Dimensional Data

Abstract: High dimensional data arises from many aspects of life and matter such as medical records, environment monitoring, business markets, social networks etc. It is a challenge for humans to understand the intricate relations among the data. Visual Analytics can offer powerful mechanisms to assist humans in the exploration and utilization of these complex data, by mining the relations from the raw data and sculpting them as visualizations associated with interaction to gain insight. All of the visual tools use some kind of projection strategy to convey the high dimensional space within the confines of the two screen dimensions. Since this projection is an inherently ill-posed problem in all but the most trivial cases, all methods will bear certain trade-offs. Knowing the strengths and weaknesses of the various paradigms existing in the field can inform the design of the most appropriate visualization strategy for the task at hand. It can help practitioners in selecting the best among the many tosols available, and it can help researchers in devising new tools to advance the state of the art.

Bio: Shenghui Cheng is a research scientist/fellow at Shenzhen Research Institute of Big Data, affiliated with Chinese University of Hong Kong, China. His research focuses on big data, data science, data visualization, visual analytics, data mining etc. He was a guest researcher at Brookhaven National Lab, USA from 2016 to 2018 and also a scientific researcher at Institute for Medical Informatics, Statistics, and Epidemiology, Leipzig University, Germany in 2015. He received his Master and PhD degree both in computer science from Stony Brook University, USA in 2016 and 2018 respectively.

Open Data Science Conference