Online Program Home
My Program

Abstract Details

Activity Number: 167
Type: Topic Contributed
Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #320864
Title: Studying the Importance of Variables for Clustering
Author(s): Volodymyr Melnykov* and Yana Melnykov and Xuwen Zhu
Companies: University of Alabama and University of Alabama and University of Alabama
Keywords: finite mixture model ; model-based clustering ; variable selection ; classification ; EM algorithm

It is well documented in literature that the choice of variables severely affects the results of cluster analysis and classification. This happens due to the fact that some variables carry important clustering information while others contain redundant or no clustering information. Several approaches to identifying variables important for clustering are proposed in recent years. We discuss a novel approach to the problem, with illustrative examples and experimental justification.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association