Table 6 Parameter settings for fair comparison.
From: A novel hybrid feature selection method combining binary grey wolf optimization and cuckoo search
Shared parameters (identical for all algorithms): | ||
|---|---|---|
Parameter | Value | Applied to |
Population size (M) | 20 | All algorithms |
Maximum iterations (T) | 150 | All algorithms |
Stopping criterion | Δfitness < 0.003 over 15 iter. | All algorithms |
Random seed | 100 | All algorithms |
KNN classifier | k = 5 | All wrapper methods |
Fitness weights | α = 0.99, β = 0.01 | All methods |
Algorithm-specific parameters: | ||
|---|---|---|
Algorithm | Parameter | Value |
BGWOCS | \(\:\lambda\:\) (Lévy flight exponent) | 1.5 |
\(\:{p}_{0}\) (Initial mutation probability) | 0.15 | |
HRO-GWO | \(\:w\) (Inertia weight) | 0.7 |
\(\:c\) (Cognitive coefficient) | 1.4 | |
GWOGA | \(\:pm\) (Mutation probability) | 0.15 |
\(\:pc\) (Crossover probability) | 0.85 | |
MTBGWO | \(\:{a}_{max}\)aximum control parameter) | 2.0 |
\(\:\theta\:\) (Threshold parameter) | 0.4 | |
IBGWO | \(\:b\) (Control parameter) | 1.2 |
\(\:r\) (Randomization factor) | 0.5 | |
