CAIP
2009
Springer
13 years 8 months ago
2009
Springer
Binary decision trees based on univariate splits have traditionally employed so-called impurity functions as a means of searching for the best node splits. Such functions use estim...