PeronaFreemanSC: Bi-parted Spectral Clustering. Peronna and Freeman.

In sClust: R Toolbox for Unsupervised Spectral Clustering

Description

Bi-parted spectral clustering based on Peronna and Freeman algorithm, which separates the data into two distinct clusters

Usage

PeronaFreemanSC(W, flagDiagZero = FALSE, verbose = FALSE)

Arguments

W	Gram Similarity Matrix.
flagDiagZero	if True, Put zero on the similarity matrix W.
verbose	To output the verbose in the terminal.

Value

returns a list containing the following elements:

• cluster: a vector containing the cluster

• eigenVect: a vector containing the eigenvectors

• eigenVal: a vector containing the eigenvalues

Author(s)

Emilie Poisson Caillault and Erwan Vincent

References

Perona, P. and Freeman, W. (1998). A factorization approach to grouping. In European Conference on Computer Vision, pages 655-670

Examples

### Example 1: 2 disks of the same size
n<-100 ; r1<-1
x<-(runif(n)-0.5)*2;
y<-(runif(n)-0.5)*2
keep1<-which((x*2+y*2)<(r1*2))
disk1<-data.frame(x+3*r1,y)[keep1,]
disk2 <-data.frame(x-3*r1,y)[keep1,]
sameTwoDisks <- rbind(disk1,disk2)
W <- compute.similarity.ZP(scale(sameTwoDisks))
res <- PeronaFreemanSC(W,flagDiagZero=TRUE,verbose=TRUE)
plot(sameTwoDisks, col = res$cluster)
plot(res$eigenVect[,1:2], col = res$cluster, main="spectral space",
     xlim=c(-1,1),ylim=c(-1,1)); points(0,0,pch='+');
plot(res$eigenVal, main="Laplacian eigenvalues",pch='+'); 

### Example 2: Speed and Stopping Distances of Cars
W <- compute.similarity.ZP(scale(iris[,-5]))
res <- PeronaFreemanSC(W,flagDiagZero=TRUE,verbose=TRUE)
plot(iris, col = res$cluster)
plot(res$eigenVect[,1:2], col = res$cluster, main="spectral space",
     xlim=c(-1,1),ylim=c(-1,1)); points(0,0,pch='+');
plot(res$eigenVal, main="Laplacian eigenvalues",pch='+'); 

sClust documentation built on Aug. 24, 2021, 1:06 a.m.