Multi-Objective Evolutionary Rule-Based Classification with Categorical Data

. 2018 Sep 7;20(9):684. doi: 10.3390/e20090684

Algorithm 4 Crowding_distance function.

Require:P {Population}
Require: P {Population}
Require: I {Individual}
Require: l {Number of objectives}

1:
for $j = 1$ to l do
2:
$f_{j}^{m a x} \leftarrow max_{I \in P} {f_{j}^{I}}$
3:
$f_{j}^{m i n} \leftarrow min_{I \in P} {f_{j}^{I}}$
4:
$f_{j}^{s u p_{j}^{I}} \leftarrow$ value of the jth objective for the individual higher adjacent in the jth objective to the individual I
5:
$f_{j}^{i n f_{j}^{I}} \leftarrow$ value of the jth objective for the individual lower adjacent in the jth objective to the individual I
6:
end for
7:
for $j = 1$ to l do
8:
if $f_{j}^{I} = f_{j}^{m a x}$ or $f_{j}^{I} = f_{j}^{m i n}$ then
9:
return ∞
10:
end if
11:
end for
12:
$C D \leftarrow 0.0$
13:
for $j = 1$ to l do
14:
$C D \leftarrow C D + \frac{f_{j}^{s u p_{j}^{I}} - f_{j}^{i n f_{j}^{I}}}{f_{j}^{m a x} - f_{j}^{m i n}}$
15:
end for
16:
return $C D$