Clustering Large Applications

Can deal with larger datasets than k-means Clustering and k-medoids Clustering by applying k-medoids Clustering on each sample from random subset of dataset.

Weakness

• efficiency depends on the sample size
• biased samples will lead to biased clustering