一、报告题目

Visual Analytics for High Dimensional Data

二、报告人

成生辉

三、报告时间

6月27日（周三）14：00

四、报告地点

软件园校区办公楼二楼202会议室

五、报告主要内容

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 tools available, and it can help researchers in devising new tools to advance the state of the art.

六、报告人简介

成生辉，香港中文大学深圳大数据研究院研究员，2013年毕业于山东大学软件学院，2018年取得纽约州立大学石溪分校博士学位。先后访问韩国纽约州立大学，德国弗里德里希·席勒大学，德国莱比锡大学，美国布鲁克海文国家实验室。他的研究方向是可视化、可视分析、数据挖掘，发表论文近30篇，包括IEEE TVCG、IEEE VIS等著名期刊会议，并担任多个著名期刊和会议的审稿人。先后在2015和2016 IEEE VIS会议上获得科学可视化和可视分析最佳海报提名奖，两次受邀在IEEE VIS做高维数据可视分析的课程导师。