Supervised vs Unsupervised learning¶

What is machine learning?¶

Algorithms that can learn from observational data and can make predictions based on it. Pretty much what your own brain does too.

The model is not given any "answers" to learn from. It must make sense of the data just given the observations themselves.
Example: group (cluster) some objects together into 2 different sets. But I don't tell you what the "right" set is for any object ahead of time.
Maybe you don't know what you're looking for - you're looking for latent variables.
Example: clustering users on a dating site based on their information and behaviour. Perhaps you'll find there are groups of people that emerge that don't conform to your known stereotypes.
Cluster movies based on their properties. Perhaps our current concepts of genre are outdated?
Analyze the test of product descriptions to find the terms that carry the most meaning for a certain category.

In supervised learing, the algorithm looks at the given observations and tries to predict the correct answers.
The model created is then used to predict the answer for new, unknown values.
Example: You can train a model for predicting car prices based on car attributes using historical sales data. That model can then predict the optimal price for new cars that haven't been sold before.

If you have a set of training data that includes the value you're trying to predict - you don't have to guess if the resulting model is good or not.
If you have enough training data, you can split it into two parts: a training set and test set.
You then train the model using only the training set.
And then measure (using r-squared or some other metric) the model's accuracy by asking it to predict values for the test set and compare that to the known, true values.