est.nearneighbor2 {Rdimtools}

Near-Neighbor Information with Bias Correction

Description

Though similar to est.nearneighbor1, authors of the reference argued that there exists innate bias in the method and proposed a non-iterative algorithm to reflect local distance information under a range of neighborhood sizes.

Usage

est.nearneighbor2(X, kmin = 2, kmax = max(3, round(ncol(X)/2)))

Arguments

X

an (n\times p) matrix or data frame whose rows are observations.

kmin

minimum neighborhood size, larger than 1.

kmax

maximum neighborhood size, smaller than p.

Value

a named list containing containing

estdim

estimated intrinsic dimension.

Author(s)

Kisung You

References

Verveer PJ, Duin RPW (1995). “An Evaluation of Intrinsic Dimensionality Estimators.” IEEE Transactions on Pattern Analysis and Machine Intelligence, 17(1), 81–86.

Examples

## create an example data with intrinsic dimension 2
X = cbind(aux.gensamples(dname="swiss"),aux.gensamples(dname="swiss"))

## acquire an estimate for intrinsic dimension
output = est.nearneighbor2(X)
sprintf("* est.nearneighbor2 : estimated dimension is %.2f.",output$estdim)