Table 1.

Statistical significance of the features and attribute occurrences on feature selection using the correlation-based method and classifiers wrappers

Attribute	p value	Correlation	KNN	SVM	NB	RBF	J48	RF
Difference of two ends	0.3367	10 ∣ 10	0 ∣ 10	10 ∣ 10	8 ∣ 10	10 ∣ 10	8 ∣ 10	10 ∣ 10
Sum of values	< 0.0001	0 ∣ 0	0 ∣ 0	7 ∣ 5	2 ∣ 1	9 ∣ 0	2 ∣ 0	3 ∣ 1
Sum of squares	< 0.0001	0 ∣ 1	1 ∣ 1	6 ∣ 7	0 ∣ 2	0 ∣ 1	3 ∣ 0	6 ∣ 0
Sum of logs	< 0.0001	0 ∣ 0	0 ∣ 2	3 ∣ 5	7 ∣ 0	0 ∣ 0	2 ∣ 1	4 ∣ 3
Arithmetic mean	0.2876	10 ∣ 1	0 ∣ 0	10 ∣ 6	6 ∣ 6	8 ∣ 9	1 ∣ 0	4 ∣ 3
Geometric mean	0.0073	0 ∣ 0	0 ∣ 0	6 ∣ 3	8 ∣ 5	0 ∣ 3	2 ∣ 0	2 ∣ 2
Population variance	0.6129	2 ∣ 1	0 ∣ 2	5 ∣ 2	7 ∣ 5	1 ∣ 0	1 ∣ 1	1 ∣ 1
Sample variance	0.6491	4 ∣ 3	0 ∣ 1	3 ∣ 3	8 ∣ 4	3 ∣ 0	1 ∣ 0	4 ∣ 3
Standard deviation	0.6491	0 ∣ 0	0 ∣ 0	4 ∣ 4	5 ∣ 5	2 ∣ 1	0 ∣ 0	0 ∣ 4
Kurtosis measure	< 0.0001	10 ∣ 10	0 ∣ 3	1 ∣ 5	1 ∣ 8	4 ∣ 6	1 ∣ 0	8 ∣ 3
Skewness measure	< 0.0001	6 ∣ 3	0 ∣ 0	3 ∣ 5	3 ∣ 8	10 ∣ 2	2 ∣ 1	3 ∣ 2
SCM	< 0.0001	0 ∣ 1	0 ∣ 3	2 ∣ 1	0 ∣ 0	0 ∣ 0	0 ∣ 0	1 ∣ 0
Energy at 0∘	0.1215	0 ∣ 0	0 ∣ 0	1 ∣ 0	3 ∣ 5	1 ∣ 1	1 ∣ 1	1 ∣ 1
Entropy at 0∘	< 0.0001	8 ∣ 3	2 ∣ 0	3 ∣ 2	0 ∣ 0	0 ∣ 0	1 ∣ 1	0 ∣ 7
Inertia at 0∘	< 0.0001	10 ∣ 10	0 ∣ 9	5 ∣ 5	8 ∣ 0	8 ∣ 9	1 ∣ 0	2 ∣ 3
Homogeneity at 0∘	< 0.0001	0 ∣ 0	2 ∣ 1	4 ∣ 1	0 ∣ 0	0 ∣ 0	2 ∣ 1	0 ∣ 4
Correlation at 0∘	< 0.0001	0 ∣ 0	1 ∣ 0	3 ∣ 3	3 ∣ 5	0 ∣ 0	1 ∣ 0	0 ∣ 2
Shade at 0∘	< 0.0001	0 ∣ 0	3 ∣ 0	0 ∣ 1	0 ∣ 0	0 ∣ 0	2 ∣ 2	3 ∣ 0
Promenance at 0∘	< 0.0001	10 ∣ 4	4 ∣ 0	0 ∣ 2	0 ∣ 0	0 ∣ 0	0 ∣ 0	1 ∣ 0
Variance at 0∘	< 0.0001	0 ∣ 0	2 ∣ 0	1 ∣ 0	0 ∣ 0	2 ∣ 3	1 ∣ 0	0 ∣ 0
IDM at 0∘	< 0.0001	9 ∣ 3	5 ∣ 5	3 ∣ 4	0 ∣ 0	6 ∣ 0	3 ∣ 3	5 ∣ 3
Energy at 45∘	0.0017	6 ∣ 1	0 ∣ 0	2 ∣ 0	4 ∣ 0	2 ∣ 0	1 ∣ 0	0 ∣ 1
Entropy at 45∘	< 0.0001	0 ∣ 0	0 ∣ 0	1 ∣ 2	0 ∣ 0	0 ∣ 0	0 ∣ 2	2 ∣ 5
Inertia at 45∘	< 0.0001	6 ∣ 1	0 ∣ 5	3 ∣ 1	0 ∣ 1	7 ∣ 5	3 ∣ 0	5 ∣ 4
Homogeneity at 45∘	< 0.0001	0 ∣ 0	1 ∣ 0	3 ∣ 1	0 ∣ 0	0 ∣ 0	0 ∣ 2	1 ∣ 1
Correlation at 45∘	< 0.0001	0 ∣ 0	0 ∣ 0	3 ∣ 5	2 ∣ 0	0 ∣ 0	0 ∣ 0	0 ∣ 3
Shade at 45∘	< 0.0001	0 ∣ 0	2 ∣ 0	1 ∣ 2	0 ∣ 0	0 ∣ 0	1 ∣ 0	2 ∣ 0
Promenance at 45∘	< 0.0001	4 ∣ 0	0 ∣ 0	1 ∣ 1	0 ∣ 0	0 ∣ 0	0 ∣ 0	0 ∣ 1
Variance at 45∘	< 0.0001	0 ∣ 0	0 ∣ 0	0 ∣ 1	0 ∣ 0	0 ∣ 0	0 ∣ 1	2 ∣ 2
IDM at 45∘	< 0.0001	10 ∣ 5	3 ∣ 0	1 ∣ 3	0 ∣ 0	3 ∣ 0	2 ∣ 0	3 ∣ 1
Energy at 90∘	0.0424	0 ∣ 2	0 ∣ 0	0 ∣ 6	2 ∣ 5	4 ∣ 5	1 ∣ 3	3 ∣ 0
Entropy at 90∘	< 0.0001	2 ∣ 0	1 ∣ 0	2 ∣ 1	0 ∣ 0	0 ∣ 0	1 ∣ 0	1 ∣ 2
Inertia at 90∘	< 0,0001	1 ∣ 1	0 ∣ 2	1 ∣ 5	1 ∣ 6	7 ∣ 4	0 ∣ 3	0 ∣ 0
Homogeneity at 90∘	< 0.0001	0 ∣ 0	1 ∣ 0	1 ∣ 3	0 ∣ 0	0 ∣ 0	1 ∣ 0	1 ∣ 0
Correlation at 90∘	< 0.0001	0 ∣ 0	5 ∣ 0	1 ∣ 4	7 ∣ 5	5 ∣ 1	0 ∣ 1	2 ∣ 4
Shade at 90∘	< 0.0001	0 ∣ 0	0 ∣ 0	0 ∣ 3	0 ∣ 0	0 ∣ 0	0 ∣ 2	1 ∣ 0
Promenance at 90∘	< 0.0001	10 ∣ 10	0 ∣ 0	1 ∣ 0	0 ∣ 0	0 ∣ 0	1 ∣ 0	1 ∣ 0
Variance at 90∘	< 0.0001	0 ∣ 3	0 ∣ 0	1 ∣ 3	0 ∣ 0	1 ∣ 1	5 ∣ 1	1 ∣ 3
IDM at 90∘	< 0.0001	9 ∣ 8	1 ∣ 3	1 ∣ 4	0 ∣ 0	5 ∣ 0	0 ∣ 0	1 ∣ 1
Energy at 135∘	0.0030	10 ∣ 3	0 ∣ 0	0 ∣ 0	6 ∣ 8	10 ∣ 3	1 ∣ 2	0 ∣ 2
Entropy at 135∘	< 0.0001	0 ∣ 0	0 ∣ 0	1 ∣ 3	0 ∣ 0	0 ∣ 0	1 ∣ 1	0 ∣ 1
Inertia at 135∘	< 0.0001	8 ∣ 5	0 ∣ 7	4 ∣ 5	2 ∣ 0	8 ∣ 4	3 ∣ 1	0 ∣ 3
Homogeneity at 135∘	< 0.0001	0 ∣ 2	0 ∣ 1	4 ∣ 4	0 ∣ 0	0 ∣ 0	2 ∣ 2	2 ∣ 3
Correlation at 135∘	< 0.0001	0 ∣ 0	0 ∣ 0	2 ∣ 5	2 ∣ 0	0 ∣ 0	0 ∣ 0	3 ∣ 0
Shade at 135∘	< 0.0001	0 ∣ 0	2 ∣ 0	1 ∣ 2	0 ∣ 0	0 ∣ 0	1 ∣ 0	0 ∣ 2
Promenance at 135∘	< 0.0001	10 ∣ 0	1 ∣ 0	1 ∣ 0	0 ∣ 0	0 ∣ 0	0 ∣ 0	1 ∣ 0
Variance at 135∘	< 0.0001	0 ∣ 0	0 ∣ 0	2 ∣ 2	0 ∣ 0	0 ∣ 0	1 ∣ 0	1 ∣ 1
IDM at 135∘	< 0.0001	8 ∣ 5	6 ∣ 2	1 ∣ 5	0 ∣ 0	4 ∣ 0	3 ∣ 7	5 ∣ 4

Occurrences are displayed in pairs x ∣y, where x and y correspond to the number of occurrences of an attribute on the tenfold cross-validation using the unbalanced and the balanced datasets, respectively