A General Hybrid Clustering Technique

Saeid Amiri, Bertrand S. Clarke, Jennifer L. Clarke, Hoyt Koepke

Research output: Contribution to journalArticle

2 Scopus citations

Abstract

Here, we propose a clustering technique for general clustering problems including those that have nonconvex clusters. For a given desired number of clusters K, we use three stages to find clusters. The first stage uses a hybrid clustering technique to produce a series of clusterings of various sizes (randomly selected). The key step in this stage is to find a K-means clustering using (Formula presented.) clusters where (Formula presented.) and then join these small clusters by using single linkage clustering. The second stage stabilizes the result of stage one by reclustering via the “membership matrix” under Hamming distance to generate a dendrogram. The third stage is to cut the dendrogram to get (Formula presented.) clusters where (Formula presented.) and then prune back to K to give a final clustering. A variant on our technique also gives a reasonable estimate for KT, the true number of clusters. We provide arguments to justify the steps in the stages of our methods and we provide examples involving simulated and published data to compare our technique with other techniques. An R library, GHC, implementing our method is available from Github.

Original languageEnglish (US)
Pages (from-to)540-551
Number of pages12
JournalJournal of Computational and Graphical Statistics
Volume28
Issue number3
DOIs
StatePublished - Jul 3 2019

Keywords

  • Consistency
  • K-mean
  • Lifetime
  • Outlier
  • Single linkage
  • t-SNE

ASJC Scopus subject areas

  • Statistics and Probability
  • Discrete Mathematics and Combinatorics
  • Statistics, Probability and Uncertainty

Fingerprint Dive into the research topics of 'A General Hybrid Clustering Technique'. Together they form a unique fingerprint.

  • Cite this