Introduction

Unsupervised learning is a paradigm in machine learning where we build models without relying on labeled training data. Until this point, we dealt with data that was labeled in some way. This means that learning algorithms can look at this data and learn to categorize them based on labels. In the world of unsupervised learning, we don't have this luxury! These algorithms are used when we want to find subgroups within datasets using some similarity metric.

One of the most common methods is clustering. You must have heard this term being used quite frequently. We mainly use it for data analysis where we want to find clusters in our data. These clusters are usually found using certain kind of similarity measure such as Euclidean distance. Unsupervised learning is used extensively in many fields, such as data mining, medical imaging, stock market analysis, computer vision, market segmentation, and so on.