13
Clustering
• There are K clusters C1,…, CK with means m1,…, mK.
•
• The least-squares error is defined as
•
•
•
• Out of all possible partitions into K clusters,
  choose the one that minimizes D.
Why don’t we just do this?
If we could, would we get meaningful objects?
D = ε    ε  || xi - mk ||   .
k=1 xi Ξ Ck
K
2