Table 5.

Critical factors from the neurofuzzy logic models, coefficient of determination (train set R² in percentage) and ANOVA parameters for training (F ratio, degree of freedom (df1: model and df2: total), and f critical value for α = 0.01) for each output.

Outputs	Submodel	Significant inputs	Train Set R2 (%)	F ratio	df1, df2	f critical
SQ	1	Fe2+	75.7	14.06	10, 47	2.72
	2	${\text{SO}}_4^{2-}$×Cl−
	3	K+
	4	${\text{NH}}_4^{+}$
PR	1	K+ × EDTA− × ${\text{SO}}_4^{2-}$	84.5	18.42	13, 44	2.56
	2	Fe2+ × ${\text{BO}}_3^{-}$
SL (cm)	1	Fe2+ × ${\text{NO}}_3^{-}$ × Cu2+	94.1	21.97	24, 33	2.40
	2	K+ × Cl−
	3	Genotype
STN	1	EDTA−	73.2	14.94	8, 49	2.89
	2	${\text{SO}}_4^{2-}$
	3	K+
	4	Genotype
BC (g)	1	Fe2+	82.7	57.90	4, 53	3.69
	2	${\text{SO}}_4^{2-}$

The most significant submodels have been bold by software.