Gene Selection for Cancer Classification using Support Vector Machines

1
k-means Clustering Algorithm

• Given a set of data points as input
• Randomly assign each point to one of the k
  clusters
• Repeat until convergence
• Calculate model of each of the k clusters
• Assign each point to the cluster with the
  closest model

2
k-means Clustering Example
3
k-means Clustering Example
Randomly assign each point to one of the clusters
4
k-means Clustering Example
Calculate center of each cluster
5
k-means Clustering Example
Calculate center of each cluster
6
k-means Clustering Example
Calculate center of each cluster
7
k-means Clustering Example
Assign each point to closest cluster center
8
k-means Clustering Example
Calculate center of each cluster
9
k-means Clustering Example
Calculate center of each cluster
10
k-means Clustering Example
Calculate center of each cluster
11
k-means Clustering Example
Assign each point to closest cluster center
12
k-means Clustering Example
Calculate center of each cluster
13
k-means Clustering Example
Calculate center of each cluster
14
k-means Clustering Example
Calculate center of each cluster
15
k-means Clustering Example
Convergence
16
Does k-means always work?
17
Does k-means always work?
18
Does k-means always work?