Table 15 Friedman test and post-hoc Nemenyi results with cliff’s delta (δ) and 95% CI.

From: A novel hybrid feature selection method combining binary grey wolf optimization and cuckoo search

Metric

Friedman p-value

BGWOCS vs. Best Competitor

Cliff’s δ

95% CI

Accuracy

p < 0.001

vs. IBGWO

δ = 0.78

[0.66, 0.87]

Features

p < 0.001

vs. MTBGWO

δ = 0.80

[0.68, 0.89]

Mean Fitness

p < 0.001

vs. GWOGA

δ = 0.76

[0.63, 0.85]

  1. δ ≥ 0.8 = large effect; critical difference (CD) = 1.24 at α = 0.05.
Back to article page