Implementation of Predictive Data Mining Techniques for Identifying Risk Factors of Early AVF Failure in Hemodialysis Patients

Table 2

The extracted rules by running “rule learner.”

	Rule model
	If DiabetesM = no and sex = female, then yes (7/19)
	If DiabetesM = no and htn = no, then yes (13/28)
	If htn = no and age = range2 [49.50–64.50], then no (10/1)
	If sex = male and Hgb = range2 [8.45–9.95], then yes (7/13)
	If age = range3 [64.50–∞] and DiabetesM = yes, then no (18/10)
	If Hgb = range3 [9.95–∞] and sex = male, then yes (8/16)
	If Hgb = range3 [9.950–∞], then no (4/0)
	If sex = male and age = range3 [64.500–∞], then no (4/2)
	If sex = female and age = range1 [−∞–49.5], then yes (2/4)
	If Hgb = range1 [−∞–8.45] and DiabetesM = yes, then no (8/4)
	If sex = female and htn = yes, then yes (1/4)
	If age = range2 [49.5–64.5] and sex = male, then no (3/1)
	If sex = male, then yes (2/4)
	Correct: 135 out of 193 training examples.