frank.err {mt}

Feature Ranking and Validation on Feature Subset

Description

Get feature ranking on the training data and validate selected feature subsets by estimating their classification error rate.

Usage

frank.err(dat.tr, cl.tr, dat.te, cl.te, cl.method="svm",
          fs.method="fs.auc", fs.order=NULL, fs.len="power2", ...)

Arguments

dat.tr

A data frame or matrix of training data. Feature ranking and classification model are carried on this data set.

cl.tr

A factor or vector of training class.

dat.te

A data frame or matrix of test data. Error rates are calculated on this data set.

cl.te

A factor or vector of test class.

cl.method

Classification method to be used. Any classification methods can be employed if they have method predict (except knn) with output of predicted class label or one component with name of class in the returned list, such as randomForest, svm, knn and lda.

fs.method

Feature ranking method. If fs.order is not NULL, it is ignored.

fs.order

A vector of feature order. Default is NULL and then the feature selection will be performed on the training data.

fs.len

The lengths of feature subsets used for validation. For details, see get.fs.len.

...

Additional parameters to fs.method or cl.method.

Value

A list with components:

cl.method

Classification method used.

fs.len

The lengths of feature subsets used for validation.

error

Error rate for each feature length.

fs.method

Feature ranking method used.

fs.order

Feature order vector.

fs.rank

Feature ranking score vector.

Author(s)

Wanchang Lin

See Also

frankvali, get.fs.len

Examples

data(abr1)
dat <- abr1$pos
x   <- preproc(dat[,110:500], method="log10")  
y   <- factor(abr1$fact$class)        

dat <- dat.sel(x, y, choices=c("1","6"))
x.1 <- dat[[1]]$dat
y.1 <- dat[[1]]$cls

idx <- sample(1:nrow(x.1), round((2/3)*nrow(x.1)), replace=FALSE) 
## construct train and test data 
train.dat  <- x.1[idx,]
train.cl   <- y.1[idx]
test.dat   <- x.1[-idx,]   
test.cl    <- y.1[-idx] 

## validate feature selection on some feature subsets
res <- frank.err(train.dat, train.cl, test.dat, test.cl, 
                 cl.method="knn", fs.method="fs.auc",  
                 fs.len="power2")
names(res)
## full feature order list
res$fs.order

## validation on subsets of feature order 
res$error

## or first apply feature selection
fs <- fs.auc(train.dat,train.cl)
## then apply error estimation for each selected feature subset
res.1 <- frank.err(train.dat, train.cl, test.dat, test.cl, 
                   cl.method="knn", fs.order=fs$fs.order,  
                   fs.len="power2")

res.1$error