relieff
Rank importance of predictors using ReliefF or RReliefF algorithm
Description
[
ranks predictors using either the ReliefF or RReliefF algorithm with
idx
,weights
] = relieff(X
,y
,k
)k
nearest neighbors. The input matrix
X
contains predictor variables, and the vector
y
contains a response vector. The function returns
idx
, which contains the indices of the most important
predictors, and weights
, which contains the weights of the
predictors.
If y
is numeric, relieff
performs
RReliefF analysis for regression by default. Otherwise, relieff
performs ReliefF analysis for classification using k
nearest
neighbors per class. For more information on ReliefF and RReliefF, see Algorithms.
Examples
Input Arguments
Output Arguments
Tips
Predictor ranks and weights usually depend on
k
. If you setk
to 1, then the estimates can be unreliable for noisy data. If you setk
to a value comparable with the number of observations (rows) inX
,relieff
can fail to find important predictors. You can start withk
=10
and investigate the stability and reliability ofrelieff
ranks and weights for various values ofk
.relieff
removes observations withNaN
values.
Algorithms
References
[1] Kononenko, I., E. Simec, and M. Robnik-Sikonja. (1997). “Overcoming the
myopia of inductive learning algorithms with RELIEFF.” Retrieved from CiteSeerX:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.4740
[2] Robnik-Sikonja, M., and I.
Kononenko. (1997). “An adaptation of Relief for attribute estimation in
regression.” Retrieved from CiteSeerX: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.34.8381
[3] Robnik-Sikonja, M., and I. Kononenko. (2003). “Theoretical and empirical analysis of ReliefF and RReliefF.” Machine Learning, 53, 23–69.
Version History
Introduced in R2010b
See Also
fscnca
| fsrnca
| knnsearch
| pdist2
| sequentialfs
| plotPartialDependence
| fsulaplacian
| fscmrmr
| fsrmrmr