Supervised and unsupervised learning are two types of Machine Learning apart from reinforcement or semi-supervised learning.
Supervised Machine Learning is when we work with datasets where the output column is already present, or to put it another way we work with labeled data.
Whereas, when we do not the output we call it unsupervised Machine Learning.
We train the model in Machine Learning, in the case of Supervised learning we already provide the output to the algorithm, and the algorithm then learns from this data and tries to find the pattern. Supervised Machine Learning can be either classification or regression type.
For Example: Classifying mails into spam and not spam, or classifying various income groups or predicting the house prices is the regression problem in supervised machine learning.
Unsupervised Learning is when we do not know the answers and want that the algorithm finds out what can be done with the data. The algorithm figures out the meaning from the data, by studying the data with the help of various algorithms
Unsupervised learning examples can be Customer segmentation, pattern recognition, text mining.
We use various unsupervised algorithms like clustering (K-means, Hierarchical Clustering), association rule, dimensionality reduction, segmentation and others.