Random Subspace Aggregation for Cancer Prediction with Gene Expression Profiles

Table 3

Function and package used in R.

Function	Package	Parameter
.test()	stats	Confidence level of the interval is 0.95. Assume two variances are equal
svm()	e1071	Choose “radial” kernel; gamma is 1/dimension; epsilon is 0.1
knn()	class	Choose
rpart()	rpart	Choose method = “class”
ada()	ada	Use decision trees as base classifiers; iteration is 50; under exponential loss, type of boosting algorithm to perform is “discrete”
ipredbagg()	ipred	Use decision trees as base classifiers; number of bootstrap replications is 25