Page 158 -

Exercises 145

4.22 Perform a resubstitution and leave-one-out estimation of the classification errors for the
three classes of cork stoppers, using the features obtained by dynamic selection.
Discuss the reliability of these estimates.

4.23 Compute the 95% confidence interval of the error for the classifier designed in
Exercise 4.1 1, using the standard formula. Perform a partition method evaluation of the
classifier, with 10 partitions, obtaining another estimate of the 95% confidence interval
of the error.
4.24 Compute the decrease of impurity in the trees shown in Figure 4.41 and Figure 4.45,
using the entropy as impurity measure.
4.25 Compute the classification matrix car vs. {mas, gla, fad} for the Breast Tissue dataset
in the tree shown in Figure 4.41. Observe its dependence on the prevalences. Compute
the linear discriminant shown in the same figure.

4.26 Using the CART approach, find decision trees that discriminate the three classes of the
CTG dataset, N, S and P, using several initial feature sets that contain the four
variability indexes ASTV, ALTV, MSTV, MLTV. Compare search times and
classification performances for the several initial feature sets.