The second type of classification problem we'll consider will be multiclass classification problem where we may have k distinct classes.

And just to remind you the other most common type of supervised

Those machine learning problems are called regression problems.

Let me back up a step further, and let's look at supervised learning more generally.

Here is one other funny thing about what would happen if we were to use linear regression for a classification problem.

And treat, therefore, the number of items I sell, as a continuous value.

Y may take on say, for value zero, one, two and three.

The inflation issue split the party.

Consider a supervised learning classification problem where you have a training set like this.

And two may mean a second type of cancer, a three may mean a third type of cancer.

You might be more familiar with the y axis, and for the purpose of this type of problem, it doesn't matter.

This problem is akin to the one we had last year.

Maybe that's what Uncle's problem is.

They don't want to let her in because there's an Israeli stamp on the Laisser Passer.

(Laughter)

Fixed!

For now, we're going to start with classification problems with just two classes zero and one.

I see no analogy between your problem and mine.

No exceptions, Harry.

Please classify these books by subject.

In this and the next few videos, I want to start to talk about classification problems, where the variable y that you want to predict is discreet valued.

Okay, your problem is solved.

The developer now resolves the problem, and for resolving the developer can choose between multiple resolutions. be fixed, meaning that the problem has actually been addressed.

So, for each of these problems, should they be treated as a classification problem, or as a regression problem?

Whales are classified as mammals.

Earlier, we talked about emails, spam classification as an example of a classification problem.

And we might be given a total of M data points.

The output layer in my example in the left has four units.

And that is challenging assumptions.

We've been talking for a while about housing prediction, and suppose you have a housing classification problem rather than a regression problem, like maybe if you have different features of a house, and you want to predict what are the odds that your house will be sold within the next six months, so that will be a classification problem.

We basically have solved problem Y by bringing to to bear our solution to problem X, or to say it another way, problem Y has been reduced to problem X.

I'd like to tell you about a problem called Anomaly Detection.

Folks, the problem is solved!

How to distribute abundance is a great problem.

That means that us terrestrials occupy a minority.

Of course. What's the point in waiting when we don't even know when a spot will open up for you!

And we looked into parametric models like Naive Bayes non parametric models.

Does it really matter? If he finds out that Gonzo is dead and not a police informant, who does he automatically blame?

Before that you'll have to store these problem reports somewhere that you can classify them, mark them, and evaluate them.

We do all the work, they get all the money, here.

Their happiness is your happiness. Their sadness is your sadness.

Do they have a common cold?

Solve this problem on your own.

Enough to cause problems.

This is a problem of his own making.