Rdimtools (version 0.3.2)

do.slpe: Supervised Locality Pursuit Embedding

Description

Supervised Locality Pursuit Embedding (SLPE) is a supervised extension of LPE that uses class labels of data points in order to enhance discriminating power in its mapping into a low dimensional space.

Usage

do.slpe(X, label, ndim = 2, preprocess = c("center", "scale", "cscale",
  "decorrelate", "whiten"))

Arguments

X

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

label

a length-\(n\) vector of data class labels.

ndim

an integer-valued target dimension.

preprocess

an additional option for preprocessing the data. Default is "center". See also aux.preprocess for more details.

Value

a named list containing

Y

an \((n\times ndim)\) matrix whose rows are embedded observations.

trfinfo

a list containing information for out-of-sample prediction.

projection

a \((p\times ndim)\) whose columns are basis for projection.

References

zheng_supervised_2006Rdimtools

See Also

do.lpe

Examples

# NOT RUN {
## generate data of 2 types with clear difference
diff = 15
dt1  = aux.gensamples(n=123)-diff;
dt2  = aux.gensamples(n=123)+diff;

## merge the data and create a label correspondingly
X      = rbind(dt1,dt2)
label  = c(rep(1,123), rep(2,123))

## compare SLPE with SLPP
out1 <- do.slpp(X, label)
out2 <- do.slpe(X, label)

## visualize
par(mfrow=c(1,2))
plot(out1$Y[,1], out1$Y[,2], main="SLPP")
plot(out2$Y[,1], out2$Y[,2], main="SLPE")
# }
# NOT RUN {
# }