The talk presents the use of Coreset, a concept proposed for geometric approximation in the area of computational geometry, for sampling big datasets for clustering purpose and a related problem called visual assessment of cluster tendency (VAT). The concept of coreset is presented. Numerical examples are given to illustrate why it can be properly used as a sample set of a big dataset. Some initial results on VAT and clustering problems are shown. Finally, some potential applications are also discussed.

BIO: Le Hong Trang currently is a lecturer in the Faculty of Computer Science and Engineering at Ho Chi Minh City University of Technology, Vietnam. He received an M.S. degree of Computer Science from School of Engineering and Technology, Vietnam National University in 2009 and PhD in Applied Mathematics from University of Lisbon with exchange to KU Leuven in 2014. 
His research line is between mathematical optimization and data analysis. Topics he is interested in are geometric optimization, sample-based approach for big data clustering and also classification with imbalanced datasets, and modern optimization models for image processing.