A Novel Self-Adaptive Harmony Search Algorithm

Table 4

The report of SPSS on the univariate (relative error) analysis of variance with two fixed factors (problem/function and HMS).

(a) Tests of between-subjects effects (dependent variable: relative error).

Source	Type III sum of squares	df	Mean square	𝐹	Sig.
Corrected	3.958^a	71	.056	11.537	.000
Model
	.301	1	.301	62.211	.000
HMS	.038	3	.013	2.632	.048
Problem	3.293	17	.194	40.092	.000
HMS * Problem	.627	51	.012	2.542	.000
Error	10.089	2088	.005
Total	14.347	2160
Corrected; Total	14.046	2159

 ^a𝑅 squared = .282 (adjusted 𝑅 squared = .257).

(b) Hochberg's multiple comparisons (dependent variable: relative error).

(I) HMS	(J) HMS	Mean difference (I − J)	Std.	Sig.
5	10	.0086	.0042	.225
	20	.0100	.0042	.101
	50	.0102	.0042	.095
10	5	−.0086	.0042	.225
	20	.0014	.0042	1.000
	50	.0015	.0042	1.000
20	5	−.0100	.0042	.101
	10	−.0014	.0042	1.000
	50	.0001	.0042	1.000
50	5	−.0102	.0042	.095
	10	−.0015	.0042	1.000
	20	−.0001	.0042	1.000

Based on observed means.