Page 313 - Artificial Intellegence_v2.0_Class_11
P. 313
Hierarchical Clustering
Hierarchical clustering builds a tree of clusters. The aim of the algorithm is to produce a tiered series of nested
clusters. Each cluster is different from every other cluster, and the objects within each cluster are mostly similar to
each other.
E F F
A D E
B D
C
C
B
A
I H
H
J I
G K K
J
G
K-Means Clustering
Out of the various clustering techniques mentioned above, the simplest and very widely used clustering algorithm is
“centroid-based clustering using K-means”.
A centroid is an imaginary or real location denoting the centre of the cluster. The K-means algorithm identifies K
number of centroids, and then assigns every data point to the nearest cluster, while trying to keep the centroids as small
as possible. The algorithm has the following steps:
Step 1: Decide the number of clusters (k)
Step 2: Select k random points from the data as centroids
Step 3: Group all the points to the nearest centroid
Step 4: Calculate the centroid of newly formed clusters
Step 5: Repeat steps 3 and 4
It is a repetitive process. It will keep on executing until there is no change in the centroids of newly formed clusters or
the maximum number of iterations are reached.
Classification & Clustering 311

