diff --git a/trees/trees.go b/trees/trees.go index ae8271e..d968b3e 100644 --- a/trees/trees.go +++ b/trees/trees.go @@ -11,6 +11,14 @@ present, so discretise beforehand (see filters) + CART (Classification and Regression Trees): + Builds a binary decision tree using the CART algorithm + using a greedy approach to find the best split at each node. + + Can be used for regression and classficiation. + Attributes have to be FloatAttributes even for classification. + Hence, convert to Integer Labels before hand for Classficiation. + RandomTree: Builds a decision tree using the ID3 algorithm by picking the Attribute amongst those