K means clustering
.It is the mostly used clustering methods. This algorithm assumes a geometric interpretation of the data sets and the datasets can be represented as some points in an n-dimensional data space. It aims to partition the n observations into k clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster. This results in a partitioning of the data space into Voronoi cells. The steps for K-means of clustering are as follows:
-
Construct a partition of a database D of n different objects into a set of k cluster centre. Here, we select the k clusters randomly.
-
Now, assign every item of the datasets to its nearest cluster centroid.
-
Recalculate the centre of the clusters and move items nearby to the clusters.
-
Repeat steps 2 and 3 until the clusters stops changing their positions and move cluster centre to cluster