A Predictive Model for Guillain–Barré Syndrome Based on Ensemble Methods

Table 2

Base classifiers used in Random Subspace for each classification case.

Base classifier	Parameter setting	Classes
kNN	k = 18, d = Manhattan	AIDP, AMAN, AMSAN, MF
kNN	k = 18, d = Manhattan	AIDP vs. ALL
kNN	k = 18, d = Manhattan	AMAN vs. ALL
kNN	k = 18, d = Manhattan	AMSAN vs. ALL
Naive bayes	—	MF vs. ALL
yjJRip	NumOpt = 3	AIDP vs. AMAN
SVMGaus	s = 0.01, C = 10	AIDP vs. AMSAN
OneR	—	AIDP vs. MF
kNN	k = 18, d = Manhattan	AMAN vs. AMSAN
SVMGaus	s = 0.01, C = 10	AMAN vs. MF
Naive bayes	—	AMSAN vs. MF