SFtools (version 0.1.0)

UfsCov: UfsCov algorithm for unsupervised feature selection

Description

Applies the UfsCov algorithm based on the space filling concept, by using a sequatial forward search (SFS).

Usage

UfsCov(data)

Arguments

data

Data of class: matrix or data.frame.

Value

A list of two elements:

• CovD a vector containing the coverage measure of each step of the SFS.

• IdR a vector containing the added variables during the selection procedure.

Details

Since the algorithm is based on pairwise distances, and according to the computing power of your machine, large number of data points can take much time and needs more memory. See UfsCov_par for parellel computing, or UfsCov_ff for memory efficient storage of large data on disk and fast access (by using the ff and the ffbase packages).

References

M. Laib and M. Kanevski (2017). Unsupervised Feature Selection Based on Space Filling Concept, arXiv:1706.08894.

Examples

infinity<-Infinity(n=800)
Results<- UfsCov(infinity)

cou<-colnames(infinity)
nom<-cou[Results[[2]]]
par(mfrow=c(1,1), mar=c(5,5,2,2))
names(Results[[1]])<-cou[Results[[2]]]
plot(Results[[1]] ,pch=16,cex=1,col="blue", axes = FALSE,
xlab = "Added Features", ylab = "Coverage measure")
lines(Results[[1]] ,cex=2,col="blue")
grid(lwd=1.5,col="gray" )
box()
axis(2)
axis(1,1:length(nom),nom)
which.min(Results[[1]])



#### UfsCov on the Butterfly dataset ####
require(IDmining)

N <- 1000
raw_dat <- Butterfly(N)
dat<-raw_dat[,-9]

Results<- UfsCov(dat)
cou<-colnames(dat)
nom<-cou[Results[[2]]]
par(mfrow=c(1,1), mar=c(5,5,2,2))
names(Results[[1]])<-cou[Results[[2]]]

plot(Results[[1]] ,pch=16,cex=1,col="blue", axes = FALSE,
xlab = "Added Features", ylab = "Coverage measure")
lines(Results[[1]] ,cex=2,col="blue")
grid(lwd=1.5,col="gray" )
box()
axis(2)
axis(1,1:length(nom),nom)
which.min(Results[[1]])