K-Means Cluster Analysis

1
Section 7.1

• K-Means Cluster Analysis

2
Objectives

• Discuss the concept of k-means clustering.
• Define measures of distance in cluster analysis.
• Understand the dangers of forced clustering.
• Generate a cluster analysis and interpret the
  results.

3
Unsupervised Classification
Training Data
Training Data
case 1 inputs, ? case 2 inputs, ? case 3
inputs, ? case 4 inputs, ? case 5 inputs, ?
case 1 inputs, cluster 1 case 2 inputs, cluster
3 case 3 inputs, cluster 2 case 4 inputs,
cluster 1 case 5 inputs, cluster 2
new case
new case
4
K-means Clustering
5
Assignment
6
Reassignment
7
Euclidean Distance
(U2,V2)
(U1,V1)
L2 ((U1 - U2)2 (V1 - V2)2)1/2
8
Manhattan Distance
(U2,V2)
(U1,V1)
L1 U1 - U2 V1 - V2
9
Forced Clustering
Last Month
Next Month
100 Cotton
60/40 Blend
60/40 Blend
10
Demonstration

• This demonstration illustrates conducting cluster
  analysis with Enterprise Miner.

11
Section 7.2

• Self-Organizing Maps

12
Objectives

• Discuss the concept of self-organizing maps.
• Generate a self-organizing map and interpret the
  results.

13
Self-Organizing Maps
Winner!
Neighbor
Neighbor
Neighbor
14
Self-Organizing Maps
Winner!
Neighbor
Neighbor
Neighbor
15
Self-Organizing Maps
16
Self-Organizing Maps
17
Self-Organizing Maps
18
Demonstration

• This demonstration illustrates generating a
  self-organizing map with Enterprise Miner.

19
Self-Organizing Map Results
younger, unmarried males unmarried males married males
younger, unmarried females unmarried females married females