What are 3 simple but important techniques in cluster analysis?

1. Density based Cluster Analysis

2. Distribution  based Cluster Analysis

3. Centroid based Cluster Analysis

Explanation:

The three important techniques of Cluster Analysis are explained below:

1. Density  based Cluster Analysis:

Density clustering involve the analysis of the clusters which are formed as dense regions in data spaces which are isolated from one another by low density regions . Moreover, the density of any cluster is higher than the density within the regions of noise.

2. Distribution  based Cluster Analysis:

In distribution clustering, we usually check that how much probability of a specific object belonging to a cluster is there.

For this we point the centroid at random and then analyse the cluster density to that point. In case, the marked point is quite far away then the probability of being in that cluster would be 0% and if it is much closer then we say that it surely belongs to that cluster.

3. Centroid based Cluster Analysis:

The representation of clusters or groups in centroid clustering is by a central vector which is not bound to be member of the data space.

The process involves selecting many crystals and then the determination of centroid at random for each and every crystal is done in order to determine that how distant is one from the cluster. If the data member is close to the centroid then it belongs to that particular cluster otherwise not.