Multi-Population Genetic Algorithm for Multilabel Feature Selection Based on Label Complementary Communication

. 2020 Aug 10;22(8):876. doi: 10.3390/e22080876

Algorithm 3 Communication function

1:
input: $P, A$ ; ▹ the sub-population P, the label-spcific accuracy matrix $A = (a_{i j}) \in R^{| P | \times | L |}$
2:
output: $i n d^{c}$ ; ▹ the complementary individual $i n d^{c}$
3:
$b \leftarrow$ find an index of the best individual in the P; ▹ the best individual $i n d_{b}$
4:
$L_{e} \leftarrow$ find an index set of labels with the highest error based on $a_{b} = [a_{b 1}, \dots, a_{b | L |}]$ ;
5:
for each individual $i n d_{i} \in P^{'}$ do ▹ the other sub-populations $P^{'}$
6:
$c_{i} \leftarrow \sum_{j \in L_{e}} a_{i j}^{'}$ ; ▹ the degree of complementarity c
7:
end for
8:
$i n d^{c} \leftarrow$ find a individual with highest c;