Skip to main content

Table 4 Average redundancy rate in terms of normalized mutual information (denoted as \(RR_{avg}^{mi}\)) of different subsets of features selected using various algorithms. Least value of \(RR_{avg}^{mi}\) signifies better choice of the feature selection algorithm

From: Comparison of five supervised feature selection algorithms leading to top features and gene signatures from multi-omics data in cancer

Dataset

Feature selection algorithm

mRMR

INMIFS

DFS

SVM-RFE-CBR

VWMRmR

Exp

0.0978

0.1053

0.0988

0.1053

0.0948a

ExpExon

0.0972

0.0991

0.0941a

0.105

0.0961

hMethyl27

0.0984

0.1

0.1045

0.0888a

0.0894

Gistic2

0.3159

0.215

0.4258

0.3209

0.1958a

Pathway activity data

0.0659

0.0533

0.0778

0.1215

0.0487

  1. aBold font justifies least \(RR_{avg}^{mi}\) among all algorithms (i.e., best performance) for each data (row)