PSO-LocBact: A Consensus Method for Optimizing Multiple Classifier Results for Predicting the Subcellular Localization of Bacterial Proteins

Table 6

PSO-LocBact configuration variables.

Configuration variable	Value type	Default value	Description
w1	Float	0.9	Inertial weight value at the beginning of PSO
w2	Float	0.4	Inertial weight value at the end of PSO
c1i	Float	2.5	Cognitive coefficient value at the beginning of PSO
c1f	Float	0.5	Cognitive coefficient value at the end of PSO
c2i	Float	0.5	Social coefficient value at the beginning of PSO
c2f	Float	2.5	Social coefficient value at the end of PSO
Particle num	Integer	25	Number of particles generated in the swarm
MAXOBJ	Integer	1,000	Maximum number of allowable objective function calls
MAXITER	Integer	—	Maximum number of allowable iterations; if this value is set, MAXOBJ will be ignored
(Program_name)	String	(Gram-negative: CELLO, PSORTb 3.0, CELLO2GO, SOSUI-GramN, SLP-Local, ngLOC, Gneg-mPLoc, PSLpred, LocTree3; Gram-positive: CELLO, PSORTb 3.0, CELLO2GO, ngLOC, Gpos-mPLoc, LocTree3)	A list of names of the programs used to calculate the final result
(Weight)	Float		A list of weights given to represent the reliability of every program included