Decision trees on interval valued variables
Chérif Mballo, Edwin Diday · 2005
In symbolic data analysis, the values of the variables can be, among others, intervals. The algebraic structure of these variables leads us to adapt dissimilarity ones to be able to study them. The Kolmogorov-Smirnov criterion is used as a test selection metric for decision tree induction. For this criterion, the values taken by the explanatory variables have to be ordered. We have been interested to different possible orders of these interval values. In this paper, we study, on the one hand, the reliability of the Kolmogorov-Smirnov criterion, and, on the other hand, we compare this criterion with the classical criterion of Gini. 1