CAIM

A Python implementation of Class-attribute interdependence maximization algorithm¹ for supervised discretization of datasets w/ missing values

Reference

[1] "L. A. Kurgan and K. J. Cios (2004), CAIM discretization algorithm, in IEEE Transactions on Knowledge and Data Engineering, vol. 16, no. 2, pp. 145-153, Feb. 2004. doi: 10.1109/TKDE.2004.1269594" http://ieeexplore.ieee.org/document/1269594

Installation

Install the following requirements:

• NumPy
• scikit-learn

pip3 install caimcaim

Example of usage

>>> from caimcaim import CAIMD
>>> from sklearn.datasets import load_iris
>>> iris = load_iris()
>>> X = iris.data
>>> y = iris.target

>>> caim = CAIMD()
>>> x_disc = caim.fit_transform(X, y)