Data clustering

In partitional clustering method the
dataset is divided into non-overlapping clusters such that each data object
belongs to exactly one cluster. Objects in one cluster are similar or close
to each other, but are dissimilar or far away from objects in other clusters.Two major
and well-known algorithms are the K-Means algorithm and the K-medoids
algorithm. In the K-Means algorithm, each cluster is represented by the
mean value of the objects in the cluster. In the K-medoids algorithm, each
cluster is represented by one of the objects located near the center of the
cluster.A threshold is used to determine if
objects will be added to existing clusters or if a new cluster is created.For example, to cluster a set of items, the first item of the set is placed in a
cluster by itself. Then, we look at the second item and based on a distance
threshold, we decide if it should be added to the first cluster or placed in
a new cluster.