You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "Pablo J. Villacorta (JIRA)" <ji...@apache.org> on 2018/07/01 13:54:00 UTC

[jira] [Created] (SPARK-24710) Information Gain Ratio for decision trees

Pablo J. Villacorta created SPARK-24710:
-------------------------------------------

             Summary: Information Gain Ratio for decision trees
                 Key: SPARK-24710
                 URL: https://issues.apache.org/jira/browse/SPARK-24710
             Project: Spark
          Issue Type: New Feature
          Components: ML
    Affects Versions: 2.3.1
            Reporter: Pablo J. Villacorta
             Fix For: 2.3.1


Spark currently uses Information Gain (IG) to decide the next feature to branch on when building a decision tree. In case of categorical features, IG is known to be biased towards features with a large number of categories. [Information Gain Ratio|https://en.wikipedia.org/wiki/Information_gain_ratio] solves this problem by dividing the IG by a number that characterizes the intrinsic information of a feature.

As far as I know, Spark has IG but not IGR. It would be nice to have the possibility to choose IGR instead of IG.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org