Enhanced Visual Analysis for Cluster Tendency Assessment and Data Partitioning
Place: Large Lecture Room - CVC
Affiliation: National Laboratory of Pattern Recognition (NLPR) Institute of Automation, Chinese Academy of Sciences. Beijing. China
Before my talk, I would like to briefly introduce our lab, and review my main research topics as well, so as to explore the possibilities of establish collaboration between CVC and NLPR. Then I will switch to a specific problem addressed in my talk, namely enhanced visual cluster analysis. Given a pairwise dissimilarity matrix of a set of objects, this talk proposes a Spectral VAT algorithm to bettwe reveal hidden data structure. A strategy for automatic determination of the number of clusters is then proposed, as well as a visual method for cluster formation. A sampling-based extended scheme is also proposed to enable visual cluster analysis for larger data sets.