6.3 Example: Titanic dataset

Let us consider the random forest model titanic_rf and passenger Johnny D as the instance of interest in the Titanic data.

Table 6.2: Variable-importance measures for the random forest model and Johnny D.

variable \(j\)	\(E_{\underline{X}}\left\{ f(\underline{X}) | \underline{X}^{J} = \underline{x}^{J}_*\right\}\)	\(\Delta^{j|J}(\underline{x}_*)\)
intercept \((v_0)\)	0.2353095	0.2353095
age = 8	0.5051210	0.2698115
class = 1st	0.5906969	0.0855759
fare = 72	0.5443561	-0.0463407
gender = male	0.4611518	-0.0832043
embarked = Southampton	0.4584422	-0.0027096
sibsp = 0	0.4523398	-0.0061024
parch = 0	0.4220000	-0.0303398
prediction	0.4220000	0.4220000

The results from Table 6.2 are presented in the following BD plot.

Source: Figure 6.3