Competitive Particle Swarm Optimization for Multi-Category Text Feature Selection

. 2019 Jun 18;21(6):602. doi: 10.3390/e21060602

Algorithm 2 Initialization function.

1:
input: $m_{c}, m_{f}$ ; ▹ The number of particles for each group $m_{c}$ , $m_{f}$
2:
output: $C, F$ ; ▹ initialized particles
3:
for $k = 1$ to $m_{c}$ do
4:
for $j = 1$ to d do
5:
$C_{k} (j) \leftarrow$ sample from $U (0, 1)$ ; ▹ use uniform distribution
6:
end for
7:
end for
8:
$X \leftarrow$ calculate a score vector; ▹ use score function of feature filter
9:
$σ \leftarrow$ calculate a standard deviation; ▹ use Equation (4)
10:
$σ \leftarrow σ m_{f}$
11:
for $k = 1$ to $m_{f}$ do
12:
for $j = 1$ to d do
13:
$F_{k} (j) \leftarrow$ sample from $N (X, σ^{2})$ ; ▹ use Gaussian distribution
14:
end for
15:
end for