Clustering

The objects in a dataset are grouped or clustered based on the principle that objects in one cluster have
high similarity to one another but are very dissimilar to objects in other clusters.In clustering data objects have no class label. That means when we start clustering we do not know what the resulted clusters will be, or by which attribute the data will be clustered.clustering is also called unsupervised learning.Before running any clustering algorithm, the data analyst removes any irrelevant meaningless attributes.