Entropy Discretization whit defined number of intervals
Entropy Discretization whit defined number of intervals
As long as i know there are more or less two aproaches to the problem of discretization: chose the number of intervals or chose the entropy of the intervals. In first case, the classes derived from the intervals will contain the same number of positive an negative examples. In the second the algorithm tries to maximize the diference between positive and negative examples.
Is there a discretization algorithm (if possible whit a python or R implementation) that allows to chose the number of intervals and then the algorithm tries to find those cutpoints that maximize de difference of pos/neg examples in each interval??
By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.
Comments
Post a Comment