Comparison of Different Machine Learning Techniques to Predict Diabetic Kidney Disease

Table 3

Confusion matrix of different classifiers.

Classifiers	Prediction		Actual state (clinical definition) (197 DKD and 213 not DKD)
	DKD	Not DKD
IBK	186	11	DKD
	15	198	NOT DKD
Random tree	186	11	DKD
	15	198	NOT DKD
Random forest	184	13	DKD
	14	199	NOT DKD
Multilayer perceptron	184	13	DKD
	15	198	NOT DKD
J48	174	23	DKD
	19	194	NOT DKD
Hoeffding tree	36	177	DKD
	81	116	NOT DKD
REP tree	171	26	DKD
	35	178	NOT DKD
Naïve bayes	165	32	DKD
	46	167	NOT DKD
AdaBoostM1	172	25	DKD
	61	152	NOT DKD