Skip to main content
Article
TiK-means: Transformation-infused K-means clustering for skewed groups
arXiv
  • Nicholas S. Berry, Iowa State University
  • Ranjan Maitra, Iowa State University
Document Type
Article
Publication Version
Submitted Manuscript
Publication Date
1-1-2019
Abstract

The K-means algorithm is extended to allow for partitioning of skewed groups. Our algorithm is called TiK-Means and contributes a K-means type algorithm that assigns observations to groups while estimating their skewness-transformation parameters. The resulting groups and transformation reveal general-structured clusters that can be explained by inverting the estimated transformation. Further, a modification of the jump statistic chooses the number of groups. Our algorithm is evaluated on simulated and real-life datasets and then applied to a long-standing astronomical dispute regarding the distinct kinds of gamma ray bursts.

Comments

This is a pre-print of the article Berry, Nicholas S., and Ranjan Maitra. "TiK-means: K -means clustering for skewed groups." arXiv preprint arXiv:1904.09609 (2019). Posted with permission.

Copyright Owner
The Authors
Language
en
File Format
application/pdf
Citation Information
Nicholas S. Berry and Ranjan Maitra. "TiK-means: Transformation-infused K-means clustering for skewed groups" arXiv (2019)
Available at: http://0-works.bepress.com.library.simmons.edu/ranjan-maitra/32/