Rdrags: Rdrags: High-dimensional feature selection via Dense Relevant...

In aydindemircioglu/Rdrags: High-Dimensional Feature Selection via Dense Relevant Attribute Group Selector

Description

Rdrags implements the high-dimensional feature selection algorithm called Dense Relevant Attribute Group Selector by Yu, Ding and Loscalzo, KDD 2008

Dense Relevant Attribute Group Selector

Usage

Rdrags(data, maxFeatures = 10, h = 0, k = 5, useKNN = TRUE,
  kernelType = "gaussian", eps = 1e-05, maxIter = 100, lambda = 0)

Arguments

data	Data to process
maxFeatures	Number of features to be selected.
h	Bandwidth, will not be used when useKNN is TRUE.
k	Neighborhood size for k-NN. Will only be used when useKNN is TRUE.
useKNN	Should k-NN be used to determine the mean distance h?
kernelType	Type of the kernel to be used, either "gaussian" or "flat"
eps	Stopping criterion for convergence.
maxIter	Maximum number of iterations
lambda	Lambda, only used when KernelType is flat

Value

List of selected features.

Rdrags functions

Rdrags

aydindemircioglu/Rdrags documentation built on May 14, 2019, 8 a.m.