Enhancing Hyperheuristics for the Knapsack Problem through Fuzzy Logic

Table 2

Percentage of times when the hyperheuristics get better or equal results than the Oracle applied to the testing set of instances.

Metrics		Methods
		HH4	HH6	HH8	FuzzyHH
Average	Equal	41.10%	39.24%	39.26%	44.67%
	Better	0.12%	0.05%	0.06%	0.20%
Standard deviation	Equal	5.13%	4.41%	5.34%	1.97%
	Better	0.19%	0.11%	0.13%	0.13%
Best	Equal	47.94%	45.58%	47.20%	47.94%
	Better	0.44%	0.44%	0.44%	0.58%
Worst	Equal	28.23%	29.55%	25.88%	41.32%
	Better	0	0	0	0

. Results in bold are the best.