# SSC: Sparse Subspace Clustering In T4cluster: Tools for Cluster Analysis

## Description

Sparse Subspace Clustering (SSC) assumes that the data points lie in a union of low-dimensional subspaces. The algorithm constructs local connectivity and uses the information for spectral clustering. SSC is an implementation based on basis pursuit for sparse reconstruction for the model without systematic noise, which solves

for column-stacked data matrix D. If you are interested in full implementation of the algorithm with sparse outliers and noise, please contact the maintainer.

## Usage

 1 SSC(data, k = 2) 

## Arguments

 data an (n\times p) matrix of row-stacked observations. k the number of clusters (default: 2).

## Value

a named list of S3 class T4cluster containing

cluster

a length-n vector of class labels (from 1:k).

algorithm

name of the algorithm.

## References

\insertRef

elhamifar_sparse_2009T4cluster

## Examples

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 ## generate a toy example set.seed(10) tester = genLP(n=100, nl=2, np=1, iso.var=0.1) data = tester$data label = tester$class ## do PCA for data reduction proj = base::eigen(stats::cov(data))$vectors[,1:2] dat2 = data%*%proj ## run SSC algorithm with k=2, 3, and 4 output2 = SSC(data, k=2) output3 = SSC(data, k=3) output4 = SSC(data, k=4) ## extract label information lab2 = output2$cluster lab3 = output3$cluster lab4 = output4$cluster ## visualize opar <- par(no.readonly=TRUE) par(mfrow=c(3,4)) plot(dat2[,1],dat2[,2],pch=19,cex=0.3,col=lab2,main="K=2:PCA") plot(data[,1],data[,2],pch=19,cex=0.3,col=lab2,main="K=2:Axis(1,2)") plot(data[,1],data[,3],pch=19,cex=0.3,col=lab2,main="K=2:Axis(1,3)") plot(data[,2],data[,3],pch=19,cex=0.3,col=lab2,main="K=2:Axis(2,3)") plot(dat2[,1],dat2[,2],pch=19,cex=0.3,col=lab3,main="K=3:PCA") plot(data[,1],data[,2],pch=19,cex=0.3,col=lab3,main="K=3:Axis(1,2)") plot(data[,1],data[,3],pch=19,cex=0.3,col=lab3,main="K=3:Axis(1,3)") plot(data[,2],data[,3],pch=19,cex=0.3,col=lab3,main="K=3:Axis(2,3)") plot(dat2[,1],dat2[,2],pch=19,cex=0.3,col=lab4,main="K=4:PCA") plot(data[,1],data[,2],pch=19,cex=0.3,col=lab4,main="K=4:Axis(1,2)") plot(data[,1],data[,3],pch=19,cex=0.3,col=lab4,main="K=4:Axis(1,3)") plot(data[,2],data[,3],pch=19,cex=0.3,col=lab4,main="K=4:Axis(2,3)") par(opar) 

T4cluster documentation built on Aug. 16, 2021, 9:07 a.m.