Principal Feature Analysis — do.pfa • Rdimtools

Principal Feature Analysis (Lu et al. 2007) adopts an idea from the celebrated PCA for unsupervised feature selection.

do.pfa(X, ndim = 2, ...)

Arguments

X

an \((n\times p)\) matrix or data frame whose rows are observations and columns represent independent variables.

ndim

an integer-valued target dimension (default: 2).

...

extra parameters including

cor: mode of eigendecomposition. FALSE for decomposing the empirical covariance matrix and TRUE uses the correlation matrix (default: FALSE).
preprocess: an additional option for preprocessing the data. See also aux.preprocess for more details (default: "center").

References

Lu Y, Cohen I, Zhou XS, Tian Q (2007). “Feature Selection Using Principal Feature Analysis.” In Proceedings of the 15th International Conference on Multimedia - MULTIMEDIA '07, 301. ISBN 978-1-59593-702-5.

Examples

# \donttest{
## use iris data
## it is known that feature 3 and 4 are more important.
data(iris)
set.seed(100)
subid    <- sample(1:150, 50)
iris.dat <- as.matrix(iris[subid,1:4])
iris.lab <- as.factor(iris[subid,5])

## compare with other methods
out1 = do.pfa(iris.dat)
out2 = do.lscore(iris.dat)
out3 = do.fscore(iris.dat, iris.lab)

## visualize
opar <- par(no.readonly=TRUE)
par(mfrow=c(1,3))
plot(out1$Y, pch=19, col=iris.lab, main="Principal Feature Analysis")
plot(out2$Y, pch=19, col=iris.lab, main="Laplacian Score")
plot(out3$Y, pch=19, col=iris.lab, main="Fisher Score")

par(opar)
# }