Desktop Survival Guide
by Graham Williams
The traditional decision tree algorithms suffer from overfitting and a bias toward selecting variables with many possible splits. The algorithms do not use any statistical significance concepts and thus, as noted by mingers:mlj.3.4.1989, cannot distinguish between significant and insignificant improvements in the information measure.
Conditional Trees (See Chapter ) take into account distributional properties.
Can library(rgl) be used to visualise a decision tree?