
Here's a quick comparison between classification and regression.

We'll go into more detail in everything that I write here.

But I want to give you a quick overview just so

you know what to look out for in this lesson.

So the first thing I'll mention is the output type of the algorithm.

As you know, for supervised classification,

this is going to be discrete in the form of class labels.

For regression this is continuous.

Which basically means that you'll be predicting a number using a regression.

The next question, a very important question is what are you actually trying to

find when you perform a classification or a regression?

In the case of classification this is usually a decision boundary.

And then depending on where a point follows relative to

that decision boundary you can assign it a class label.

With a regression what we're trying to find is something we usually call

a best fit line.

Which is a line that fits the data rather than a boundary that describes it.

This will make a lot more sense in the next couple of videos.

The last thing I'll mention is how to evaluate.

When we were doing supervised classification we usually use the accuracy,

which is whether it got the class labels correct or not on your test set.

And for regression we'll talk about a couple different evaluation metrics.

One of which is called the sum of a squared error.

Another one is called r squared.

I'll tell you a lot more about both of these in the middle part of the lesson.

The main point I want you to take away from this is that,

while regression isn't exactly the same as supervised classification.

A lot of the things you already know about supervised classification have direct

analogs in regression.

So you should think of regression as a different type of supervised learning.

Not as a completely new topic that you now have to learn from scratch.