One of the simplest and most powerful techniques used in unsupervised learning is based on grouping similar patterns together through clustering algorithms. It is used to understand a particular aspect of the data that is related to the problem we are trying to solve. Clustering algorithms look for natural grouping in data items. As the group is not based on any target or assumptions, it is classified as an unsupervised learning technique.
Groupings created by various clustering algorithms are based on finding the similarities between various data points in the problem space. The best way to determine the similarities between data points will vary from problem to problem and will depend on the nature of the problem we are dealing with. Let's look at the various methods that can be used to calculate the similarities between various...
