User-Oriented Summaries Using a PSO Based Scoring Optimization Method

. 2019 Jun 22;21(6):617. doi: 10.3390/e21060617

Algorithm 2: Proposed PSO algorithm

Input: popSize, maxIte

1:
initialize $φ 1$ , $φ 2$ , $φ 3$ and $φ 4$
2:
initialize $l i m i t 1_{j, u p p e r}$ , $l i m i t 1_{j, l o w e r}$ , $l i m i t 2_{j, u p p e r}$ and $l i m i t 2_{j, l o w e r}$ $\forall j = 1 . . n$
3:
initialize $l i m i t 3_{j, u p p e r}$ , $l i m i t 3_{j, l o w e r}$ , $l i m i t 4_{j, u p p e r}$ , $l i m i t 4_{j, l o w e r}$ $\forall j = 1 . . n$
4:
create swarm population with size maxIte
5:
repeat
6:
adjust inertia factor value according to Equation (9)
7:
for all $i = 1 . . p o p S i z e$ do
8:
$f i t I n d_{i} \leftarrow$ calculate particle $p_{i}$ fitness according to Equation (10)
9:
if $f i t I n d_{i} > f i t B e s t I n d_{i}$ then
10:
$f i t B e s t I n d_{i} = f i t I n d_{i}$
11:
end if
12:
end for
13:
for all $i = 1 . . p o p S i z e$ do
14:
retrieve best particle from neighborhood
15:
update speed according to Equations (1)–(4) and (7) to binary and continuous part respectively
16:
modify particle $p_{i}$ position according to Equations (6) and (8)
17:
end for
18:
until reaching $m a x I t e$ iterations
19:
return solution of best particle in population

Output: The particle with the last best fitness value