kmedoids

Partition data into clusters using medoids.

📝Syntax

idx = kmedoids(X, k)
[idx, C, sumd, D, midx, info] = kmedoids(...)

📄Description

kmedoids clusters rows of X using a serial PAM-style update loop and Nelson's random generator for random starts.

💡Examples

Cluster observations and return medoid information.

X = [0; 1; 10; 11];
[idx, C, sumd, D, midx, info] = kmedoids(X, 2, 'Start', [1; 3])

Use cityblock distance for two-dimensional observations.

X = [0 0; 0 1; 5 5; 5 6];
[idx, C] = kmedoids(X, 2, 'Start', [0 0; 5 5], 'Distance', 'cityblock')

Used Functions

kmeans pdist2 statset rng