KMeans clustering
KMeans clustering is a method of cluster analysis which aims to
partition n observations into k clusters in which each observation
belongs to the cluster with the nearest mean. Its extension to
Soft-KMeans allows to have soft assignment to the clusters, which in
turns allows fuzzy boundaries.
More information on Wikipedia.
Parameters:
- Clusters: number of clusters to optimize
- Method: choice of hard assignment (KMeans) or soft ones (Soft-KMeans)
- Metric: type of metric to be used
- L0: infinite norm
- L1-p: norm of degree 1-p
- Power: power for Lp norm
- Beta: boundary stiffness (Soft KMeans)