(New page: Impurity of a class is defined as the amount of data misclassified into that class. It is zero when all training data belongs to one class. See Lecture 21 - Decision Trees(Continued) f...) |
(No difference)
|
Latest revision as of 10:17, 7 April 2008
Impurity of a class is defined as the amount of data misclassified into that class. It is zero when all training data belongs to one class. See Lecture 21 - Decision Trees(Continued)_OldKiwi for the definitions of different types of impurities