Classification

linear regression for classification problems.

For classification, we know that

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

Here are some examples of classification problems.

Today we'll talk about classification versus regression.

Tagging is a cooperative infrastructure answer to classification.

Our example so far was one of classification.

Regression problems are fundamentally different from classification problems.

Let's say we're using the tumor classification example.

So, how do we develop a classification algorithm?

The closest classification it matches is a virus.

In HarvardYenching, it's a classification for Asian literature.

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

We're going to consider two types of classification problems.

Now, in 1970s, Bautz and Morgan introduced a different classification.

That we could create a classification of the mental illnesses.

We're testing everyone with a level four classification or above.

But what you find is this is different from classification as before.

Support vector machines are supervised learning models used for classification and regression analysis.

This is from the classification case, except this time the Ys are continuous.

Here's our linear separation classification function where this is our common linear function.

I'm going to focus on the application of neural networks to classification problems.

We talked about classification and regression and different methods to do supervised learning.

Let's now do classification and see what class and message it ends up with.

You can toss them easily into the naive base model and get better classification.

If this is larger or equal to zero, then we call our classification 1.

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

Logistic Regression is actually a classification algorithm that we apply to settings where the

The first is binary classification, where the labels y are either zero or one.

To introduce a bit more terminology this is an example of a classification problem.

How about if we were doing a classification problem and say using logistic regression instead.

So, applying linear regression to a classification problem usually isn't, often isn't a great idea.

This is called a multi class classification problem, but for the next few videos, let's start with the two class or the binary classification problem. and we'll worry about the multi class setting later.

And so it's anachronistically a classification by fathers' occupations, so single parents go on their own.

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

We've seen classification before for spam and other types of categories, language ID and so on.

And we talked about the classification problem, where the goal is to predict a discrete value output.

Now since this is a classification task, we know that Y must be either zero or one, right?

In classification problems there is another way to plot this data. Let me show you what I mean.

And because there's a small number of discrete values, I would therefore treat it as a classification problem.

Now, you may be surprised to learn that the gzip command is actually pretty good as a classification algorithm.

levels of trust, obesity, mental illness which in standard diagnostic classification includes drug and alcohol addiction and social mobility.

looks like linear regression is actually doing something reasonable even though this is a classification task we're interested in.

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