Skip to main content

Advertisement

Table 7 The (c/ s2) error bound results of random forest models against the number of genes per leaf nmin on the ten gene data sets.

From: Genome-wide association data classification and SNPs selection using two-stage quality-based Random Forests

Data set Model n min
   1 2 5 8 10 15
COL RF .044 .032 .033 .032 .035 .034
  wsRF .046 .039 .042 .040 .042 .040
  ts-RF .053 .043 .044 .043 .046 .044
SRB RF .018 .019 .017 .017 .019 .019
  wsRF .012 .013 .013 .013 .012 .013
  ts-RF .013 .013 .011 .010 .013 .012
LEU RF .040 .037 .037 .037 .039 .039
  wsRF .035 .027 .028 .029 .032 .030
  ts-RF .023 .020 .021 .021 .022 .022
LYM RF .019 .013 .012 .012 .016 .014
  wsRF .011 .010 .010 .010 .010 .010
  ts-RF .008 .005 .005 .005 .007 .006
BR2 RF .034 .034 .033 .035 .041 .037
  wsRF .038 .039 .038 .040 .042 .041
  ts-RF .046 .039 .039 .039 .048 .045
BR3 RF .068 .056 .056 .054 .064 .057
  wsRF .065 .057 .057 .056 .059 .058
  ts-RF .086 .064 .065 .062 .076 .066
NCI RF .037 .023 .024 .025 .030 .027
  wsRF .016 .017 .016 .017 .017 .017
  ts-RF .044 .022 .022 .025 .031 .025
BRA RF .045 .030 .029 .029 .028 .031
  wsRF .024 .025 .024 .024 .024 .025
  ts-RF .041 .022 .022 .022 .021 .024
PRO RF .041 .034 .033 .032 .037 .034
  wsRF .038 .033 .032 .031 .034 .032
  ts-RF .043 .033 .032 .032 .038 .033
ADE RF .080 .073 .071 .072 .076 .075
  wsRF .068 .064 .065 .065 .065 .066
  ts-RF .054 .049 .048 .048 .051 .051
  1. Numbers in bold are the best results.