What is ML

In order to fully understand the code in this course, we recommend you take the following courses (unless you are already familiar with these topics):

Machine learning (ML) is an approach to programming in which computers learn from data to solve a task rather than having the solution explicitly programmed.

Let's look at the example of the spam/ham (not spam) classifier.

If you try to build it using the traditional programming approach (without ML), you will have a hard time writing the logic of your program, even manually creating a spam word list.

Alternatively, you can feed many examples of spam mail and ham mail to a machine learning model that will learn by itself.

The data we give to the ML model to train on is called the training set. In the example above, the training set is a bunch of emails that are already labeled as spam or ham. This allows the model to learn the characteristics of spam and non-spam emails.

After the model has been trained, we evaluate its performance using the test set. This consists of a separate set of emails, also labeled as spam or ham, which helps us determine how well our model can generalize to new, unseen data.