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

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

Algorithm 2 Initialization function

1:
input:m; ▹ the number of sub-populations m
2:
output: $P_{1}, \dots, P_{k}, \dots, P_{m}$ ; ▹ the initial sub-populations $P_{1}, \dots, P_{k}, \dots, P_{m}$
3:
for each feature $f_{i} \in F$ do ▹ the original feature set F
4:
if $H (f_{i}) = 0$ then
5:
$F \leftarrow F \ f_{i}$ ;
6:
end if
7:
end for
8:
$[C_{1}, \dots, C_{k}, \dots, C_{m}] \leftarrow$ patition F into m clusters; ▹ use the spherical k-means algorithm
9:
for $k = 1$ to m do
10:
for each individual $i n d_{i} \in P_{k}$ do
11:
$i n d_{i} \leftarrow$ initialize by selecting n features via stochastic sampling; ▹ use Equation (1)
12:
end for
13:
end for