Figure 4.

Pareto fronts of one execution of ENORA and NSGA-II, with $M_{max}=12$, on the Monk’s Problem 2 dataset, and their respective HVR. Note that in the case of multi-objective classification where ${\mathcal{F}}_{\mathcal{D}}$ is maximized (${\mathcal{ACC}}_{\mathcal{D}}$ and ${\mathcal{AUC}}_{\mathcal{D}}$), function ${\mathcal{F}}_{\mathcal{D}}$ has been converted to minimization for a better understanding of the Pareto front.