Research Article
A Novel Ensemble Credit Scoring Model Based on Extreme Learning Machine and Generalized Fuzzy Soft Sets
Table 7
Classification outcomes from all data sets for different performance measures with various combinations.
| Data set | Performance measurement | Other feature selection methods with GFSS-based combination | Traditional combination methods with AEnet-based feature selection | EGHE | Cost-sensitive | GA | IGR | Enet | WAVG | MajVot | WVOT | FSS |
| Germany | AUC | 0.786 | 0.792 | 0.777 | 0.781 | 0.777 | 0.786 | 0.789 | 0.771 | 0.823 | HM | 0.286 | 0.177 | 0.226 | 0.238 | 0.222 | 0.285 | 0.246 | 0.305 | 0.325 | BS | 0.166 | 0.208 | 0.197 | 0.182 | 0.181 | 0.183 | 0.192 | 0.157 | 0.184 | ACC | 0.842 | 0.862 | 0.833 | 0.827 | 0.845 | 0.859 | 0.836 | 0.844 | 0.886 | Australia | AUC | 0.921 | 0.927 | 0.933 | 0.928 | 0.921 | 0.922 | 0.932 | 0.928 | 0.945 | HM | 0.667 | 0.519 | 0.573 | 0.628 | 0.637 | 0.637 | 0.632 | 0.653 | 0.659 | BS | 0.092 | 0.145 | 0.173 | 0.103 | 0.101 | 0.112 | 0.107 | 0.097 | 0.095 | ACC | 0.872 | 0.878 | 0.882 | 0.863 | 0.872 | 0.873 | 0.876 | 0.874 | 0.895 | Japan | AUC | 0.923 | 0.928 | 0.918 | 0.932 | 0.918 | 0.927 | 0.928 | 0.925 | 0.937 | HM | 0.650 | 0.492 | 0.566 | 0.618 | 0.602 | 0.642 | 0.617 | 0.648 | 0.660 | BS | 0.099 | 0.157 | 0.174 | 0.109 | 0.121 | 0.112 | 0.116 | 0.103 | 0.098 | ACC | 0.871 | 0.862 | 0.861 | 0.868 | 0.853 | 0.864 | 0.873 | 0.868 | 0.904 | Iran | AUC | 0.791 | 0.778 | 0.781 | 0.767 | 0.777 | 0.779 | 0.783 | 0.778 | 0.802 | HM | 0.279 | 0.150 | 0.217 | 0.162 | 0.283 | 0.108 | 0.107 | 0.294 | 0.303 | BS | 0.042 | 0.069 | 0.056 | 0.049 | 0.044 | 0.048 | 0.049 | 0.044 | 0.059 | ACC | 0.883 | 0.881 | 0.874 | 0.861 | 0.867 | 0.885 | 0.877 | 0.883 | 0.915 | Bene 1 | AUC | 0.843 | 0.812 | 0.822 | 0.821 | 0.882 | 0.879 | 0.886 | 0.888 | 0.881 | HM | 0.396 | 0.263 | 0.338 | 0.258 | 0.324 | 0.441 | 0.385 | 0.447 | 0.351 | BS | 0.161 | 0.240 | 0.256 | 0.197 | 0.188 | 0.159 | 0.198 | 0.154 | 0.173 | ACC | 0.881 | 0.872 | 0.864 | 0.872 | 0.872 | 0.872 | 0.865 | 0.876 | 0.896 | Bene 2 | AUC | 0.921 | 0.838 | 0.858 | 0.868 | 0.878 | 0.844 | 0.876 | 0.883 | 0.888 | HM | 0.537 | 0.401 | 0.478 | 0.489 | 0.505 | 0.436 | 0.432 | 0.496 | 0.506 | BS | 0.091 | 0.136 | 0.168 | 0.122 | 0.102 | 0.115 | 0.112 | 0.103 | 0.114 | ACC | 0.878 | 0.860 | 0.857 | 0.868 | 0.881 | 0.875 | 0.882 | 0.884 | 0.898 | Shuttle | AUC | 0.896 | 0.914 | 0.914 | 0.922 | 0.897 | 0.919 | 0.917 | 0.906 | 0.943 | HM | 0.653 | 0.510 | 0.562 | 0.625 | 0.620 | 0.635 | 0.622 | 0.636 | 0.658 | BS | 0.091 | 0.143 | 0.170 | 0.103 | 0.099 | 0.110 | 0.104 | 0.096 | 0.095 | ACC | 0.852 | 0.866 | 0.862 | 0.858 | 0.849 | 0.870 | 0.862 | 0.851 | 0916 | Skin_segment | AUC | 0.902 | 0.913 | 0.899 | 0.926 | 0.896 | 0.922 | 0.915 | 0.903 | 0.936 | HM | 0.637 | 0.486 | 0.555 | 0.615 | 0.586 | 0.638 | 0.607 | 0.631 | 0.659 | BS | 0.096 | 0.154 | 0.171 | 0.107 | 0.119 | 0.110 | 0.113 | 0.098 | 0.098 | ACC | 0.853 | 0.849 | 0.844 | 0.863 | 0.833 | 0.862 | 0.859 | 0.845 | 0.908 | MiniBooNE | AUC | 0.773 | 0.767 | 0.764 | 0.763 | 0.757 | 0.775 | 0.770 | 0.758 | 0.800 | HM | 0.272 | 0.149 | 0.213 | 0.159 | 0.277 | 0.109 | 0.104 | 0.286 | 0.302 | BS | 0.042 | 0.067 | 0.054 | 0.048 | 0.0439 | 0.047 | 0.047 | 0.041 | 0.059 | ACC | 0.875 | 0.877 | 0.885 | 0.876 | 0.884 | 0.880 | 0.893 | 0.890 | 0.903 | LC2017Q1 | AUC | 0.823 | 0.801 | 0.804 | 0.816 | 0.859 | 0.874 | 0.872 | 0.867 | 0.879 | HM | 0.386 | 0.260 | 0.332 | 0.257 | 0.315 | 0.440 | 0.378 | 0.435 | 0.350 | BS | 0.158 | 0.237 | 0.252 | 0.195 | 0.182 | 0.157 | 0.196 | 0.149 | 0.203 | ACC | 0.861 | 0.888 | 0.877 | 0.875 | 0.871 | 0.878 | 0.871 | 0.873 | 0.912 |
|
|