K-Medoids clustering

K-Means variation for object-distance array

(comments in French within the implemented methods)

Outline

• Benchmark with Adjusted Rand Index computed with adj.rand.index function from package fossil
• PCA with factoextra
• Comparison with
  • PAM algorithm from cluster R package
  • k-means algorithm from stats R package
  • Mclust algorithm from mclust R package
• Complete example on iris dataset

Freely inspired by the research paper : A simple and fast algorithm for K-medoids clustering, Hae-Sang Park, Chi-Hyuck Jun <external link>

License

Lucas RODRIGUEZ (2021)