O'Reilly logo

Ensemble Machine Learning by Ankit Dixit

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Weak classifier

A weak classifier is nothing but a threshold value that can divide the data into two classes. Experiments show that the error rate must be less than 0.5; that is, we should get an accuracy of the classification more than 50%. Now, why is this known as a weak classifier? The answer is simple. Because with a single threshold value, we can’t get a very high accuracy. So I have a question for you: how to pick a threshold value that can assure you maximum accuracy? Did we have done something like that earlier? Yes we did! Don't you remember? Gini index! Yes, we have tested and picked a threshold value for a perfect split. And we will do the same again; we will pick a threshold and test whether it can be our weak classifier or not. ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required