Raw File
Tip revision: 583e8e137abae0107d3e0ea9b5549b3d141ba1c5 authored by Piotr Romanski on 11 April 2009, 00:00:00 UTC
version 0.17
Tip revision: 583e8e1
\title{ RReliefF filter }
  The algorithm finds weights of continous and discrete attributes basing on a distance between instances.
relief(formula, data, neighbours.count = 5, sample.size = 10)
  \item{formula}{ a symbolic description of a model }
  \item{data}{ data to process }
  \item{neighbours.count}{ number of neighbours to find for every sampled instance }
  \item{sample.size}{ number of instances to sample }
  The algorithm samples instances and finds their nearest hits and misses. Considering that result, it evaluates weights of attributes.
    \item{-}{Igor Kononenko: Estimating Attributes: Analysis and Extensions of RELIEF. In: European Conference on Machine Learning, 171-182, 1994.}
    \item{-}{Marko Robnik-Sikonja, Igor Kononenko: An adaptation of Relief for attribute estimation in regression. In: Fourteenth International Conference on Machine Learning, 296-304, 1997.}
a data.frame containing the worth of attributes in the first column and their names as row names
\author{ Piotr Romanski }
  weights <- relief(Species~., iris, neighbours.count = 5, sample.size = 20)
  subset <- cutoff.k(weights, 2)
  f <- as.simple.formula(subset, "Species")
back to top