Supervised and Unsupervised Learning are two Machine Learning Models. Both these models are used for data analysis. The key difference between these two models is labelled data. Supervised Learning requires labelled data whereas unsupervised Learning does not need it.
As the name suggests, supervised learning requires human supervision. This model needs labelled data to work. Experts label the data based on past available information. These models can predict outputs for inputs based on labelled data available.
Sales prediction on different holidays at a store
Predicting stock market trends
Unsupervised Learning does not human intervention. The model learns by itself based on the raw data provided to it. It tries to find patterns in the data and predicts the output. The results may not be accurate since they are not based on already established data patterns, but they will likely improve over time. Clustering is a way of unsupervised learning. In clustering, the model tries to find similarities in data and cluster them into groups.
The section 'Customers who bought this product also bought these products on e-commerce sites.
Predicting the students who are likely to join Numpy Ninja from the students who took trial classes.
Which model to use?
The model is chosen based on the problem and availability of resources. A supervised learning model might be the right approach if labelled data is available. But in many real-time scenarios, the problem at hand is more complex to stick to a single approach. The problem is broken into parts and different models may be applied at different steps.