scores: mRMR Scores as per the MI gain for each feature

In mRMRe: Parallelized Minimum Redundancy, Maximum Relevance (mRMR)

mRMR Scores as per the MI gain for each feature

Description

The scores method returns the scores of individual features in respect to previously selected features as per standard mRMR procedure. For each target, the score of a feature is defined as the mutual information between the target and this feature minus the average mutual information of previously selected features and this feature.

Usage

## S4 method for signature 'mRMRe.Data'
scores(object, solutions)
## S4 method for signature 'mRMRe.Filter'
scores(object)
## S4 method for signature 'mRMRe.Network'
scores(object)

Arguments

object	a mRMRe.Data, mRMRe.Filter or mRMRe.Network object.
solutions	a set of solutions from mRMRe.Filter or mRMRe.Network to be used in computing the scores from a mRMRe.Data set.

Author(s)

Nicolas De Jay, Simon Papillon-Cavanagh, Benjamin Haibe-Kains

Examples

set.thread.count(2)
data(cgps)
feature_data <- mRMR.data(data =  data.frame(cgps.ge))

# Create an mRMR filter and obtain the indices of selected features
filter <- mRMR.classic("mRMRe.Filter", data = feature_data, target_indices = 3:5,
						feature_count = 2)
scores(filter)

mRMRe documentation built on April 25, 2023, 9:13 a.m.