Package org.tribuo.classification.fs

package org.tribuo.classification.fs
Information theoretic feature selection algorithms.

For an overview of the field of information theoretic feature selection see: 

 Brown G, Pocock A, Zhao M-J, Lujan M.
 "Conditional Likelihood Maximisation: A Unifying Framework for Information Theoretic Feature Selection"
 Journal of Machine Learning Research (JMLR) PDF.
  • Classes
    Class
    Description
    CMIM
    Selects features according to the Conditional Mutual Information Maximisation algorithm.
    JMI
    Selects features according to the Joint Mutual Information algorithm.
    MIM
    Selects features according to their mutual information with the class label (aka Mutual Information Maximisation).
    mRMR
    Selects features according to the Minimum Redundancy Maximum Relevance algorithm.