|
I'm looking for a set of weak classifiers that work with Adaboost to test on popular datasets. Most of the examples on the web use some kind of random weak learners which work on their own randomly generated dataset. Could you point me to any usable weak learners? |
||||
|
|
|
The most basic and most common weak learner is a decision stump which is basically a single level decision tree. That means, say the points in your dataset are N dimensional (features), a decision stump is a threshold on a single dimension. Below the threshold is one class, above it the other. Pretty much any classifier could be used as a weak classifier. I have seen papers where people use more complicated (multiple levels) decision trees or Support Vector Machines as weak learners. |
|||
|
|
