Table 2.

Supervised learning results on the right eye.

Scheme	Recall	FP rate	Precision	F-measure	MCC	Correctly classified
ZeroR class 1	1.0	1.0	0.775	0.873	NC
ZeroR class 2	0.0	0.0	NC	NC	NC
ZeroR class 3	0.0	0.0	NC	NC	NC
ZeroR weighted avg	0.775	0.775	NC	NC	NC	77.494%
NaiveBayes class 1	1.0	0.522	0.876	0.934	0.650
NaiveBayes class 2	0.469	0.0	1.0	0.638	0.660
NaiveBayes class 3	0.417	0.0	1.0	0.588	0.640
NaiveBayes weighted avg	0.888	0.410	0.902	0.871	0.650	88.837%
Random forest class 1	1.0	0.010	0.997	0.999	0.993
Random forest class 2	0.984	0.0	1.0	0.992	0.990
Random forest class 3	1.0	0.0	1.0	1.0	1.0
Random forest class weighted avg	0.998	0.008	0.998	0.998	0.993	99.768%
Multilayer perceptron class 1	0.994	0.072	0.979	0.987	0.940
Multilayer perceptron class 2	0.967	0.005	0.967	0.967	0.962
Multilayer perceptron class 3	0.792	0.0	1.0	0.884	0.884
Multilayer perceptron weighted avg	0.979	0.057	0.979	0.979	0.941	97.912%
Randomizable filtered classifier + IBk class 1	1.0	0.0	1.0	1.0	1.0
Randomizable filtered classifier + IBk class 2	1.0	0.0	1.0	1.0	1.0
Randomizable filtered classifier + IBk class 3	1.0	0.0	1.0	1.0	1.0
Randomizable filtered classifier + IBk weighted avg	1.0	0.0	1.0	1.0	1.0	100%

FP false positive, MCC Matthews correlation coefficient, avg. average, NC not calculable.