Abstract :
[en] This paper focuses on the variance introduced by the discretization techniques used to handle continuous attributes in decision tree induction. Different discretization procedures are first studied empirically, then means to reduce the discretization variance are proposed. The experiments shows that discretization variance is large and that it is possible to reduce it significantly without notable computational costs. The resulting variance reduction mainly improves interpretability and stability of decision trees, and marginally their accuracy.
Scopus citations®
without self-citations
15