Skip to main content

Table 11 The obtained results for G-mean from boosting algorithms using decision tree as a base learner

From: Boosting methods for multi-class imbalanced data classification: an experimental review

Dataset Algorithm
AdaBoost.MH SAMME CatBoost LogitBoost GradientBoost XGBoost MEBoost SMOTEBoost RUSBoost LightGBM AdaC1 AdaC2 AdaC3 AdaCost
Wine 0.9058 0.9149 0.9768 0.9565 0.9908 0.9665 0.9832 0.9501 0.962 0.9688 0.9325 0.9193 0.8776 0.9347
± 0.0170 ± 0.0125 ± 0.0058 ± 0.0102 ± 0.0026 ± 0.1160 ± 0.0077 ± 0.0165 ± 0.0091 ± 0.0138 ± 0.0173 ± 0.0171 ± 0.0133 ± 0.0132
Hayes-Roth 0.7948 0.7829 0.7558 0.7901 0.7915 0.786 0.7896 0.7823 0.7793 0.3203 0.7896 0.8003 0.8059 0.7982
± 0.0179 ± 0.0116 ± 0.0129 ± 0.0110 ± 0.0162 ± 0.0156 ± 0.0194 ± 0.0164 ± 0.0225 ± 0.1187 ± 0.0127 ± 0.0129 ± 0.0225 ± 0.0204
Contraceptive 0.5243 0.4841 0.5052 0.4819 0.4849 0.5335 0.4971 0.4704 0.4327 0.4926 0.5314 0.1711 0.3828 0.539
± 0.0048 ± 0.0079 ± 0.0055 ± 0.0085 ± 0.0101 ± 0.0080 ± 0.0063 ± 0.0126 ± 0.0081 ± 0.0064 ± 0.0051 ± 0.0240 ± 0.0129 ± 0.0088
Pen-based 0.9006 0.9576 0.9615 0.9552 0.9509 0.932 0.9724 0.9496 0.9586 0.9524 0.9031 0.9035 0.9023 0.9027
± 0.0057 ± 0.0056 ± 0.0046 ± 0.0051 ± 0.0031 ± 0.0055 ± 0.0038 ± 0.0027 ± 0.0048 ± 0.0020 ± 0.0101 ± 0.0053 ± 0.0063 ± 0.0081
Vertebral column 0.7678 0.7669 0.7921 0.7696 0.7704 0.7733 0.7713 0.7907 0.7765 0.7835 0.7815 0.7801 0.7674 0.7318
± 0.0217 ± 0.0239 ± 0.0245 ± 0.0163 ± 0.0193 ± 0.0159 ± 0.0191 ± 0.0239 ± 0.0215 ± 0.0252 ± 0.0113 ± 0.0166 ± 0.0146 ± 0.0184
New thyroid 0.8971 0.8904 0.9099 0.9131 0.9244 0.9065 0.2727 0.9055 0.9131 0.9156 0.8979 0.8869 0.8389 0.8852
± 0.0176 ± 0.02311 ± 0.0106 ± 0.0140 ± 0.0098 ± 0.0176 ± 0.01659 ± 0.0240 ± 0.01361 ± 0.0091 ± 0.0268 ± 0.0327 ± 0.0322 ± 0.0238
Dermatology 0.9555 0.9625 0.9786 0.9665 0.9713 0.9693 0.8723 0.9653 0.9433 0.9000 0.9566 0.9602 0.9522 0.9546
± 0.0067 ± 0.0040 ± 0.0039 ± 0.0096 ± 0.0050 ± 0.0125 ± 0.0854 ± 0.0061 ± 0.0109 ± 0.0097 ± 0.0070 ± 0.0065 ± 0.0103 ± 0.0075
Balance scale 0.2461 0.4344 0.2898 0.2381 0.0378 0.0098 0.2856 0.4161 0.4877 0.6372 0.073 0.5198 0.4031 0.5881
± 0.0632 ± 0.0731 ± 0.1676 ± 0.0515 ± 0.0650 ± 0.0310 ± 0.0730 ± 0.0544 ± 0.1068 ± 0.0641 ± 0.0521 ± 0.0074 ± 0.0447 ± 0.0196
Glass 0.3097 0.4095 0.4591 0.5517 0.3131 0.4252 0.2413 0.2414 0.2014 0.0000 0.3252 0.3382 0.2782 0.3474
± 0.1445 ± 0.0571 ± 0.1209 ± 0.1101 ± 0.1880 ± 0.1218 ± 0.0957 ± 0.1493 ± 0.1506 ± 0.0000 ± 0.1040 ± 0.1197 ± 0.1381 ± 0.1161
Heart 0.0000 0.0000 0.0119 0.0000 0.0000 0.0000 0.0000 0.0993 0.0058 0.0000 0.0173 0.0000 0.0307 0.0066
± 0.0000 ± 0.0000 ± 0.0250 ± 0.0000 ± 0.0000 ± 0.0000 ± 0.0000 ± 0.0801 ± 0.0185 ± 0.0000 ± 0.0282 ± 0.0000 ± 0.0324 ± 0.0210
Car evaluation 0.9878 0.9474 0.9861 0.9967 0.9852 0.8401 0.9920 0.9304 0.8219 0.9882 0.603 0.8337 0.8852 0.8686
± 0.0091 ± 0.0065 ± 0.0045 ± 0.0036 ± 0.0081 ± 0.0154 ± 0.0042 ± 0.0072 ± 0.0202 ± 0.0036 ± 0.0377 ± 0.0060 ± 0.0089 ± 0.0103
Thyroid 0.9771 0.9779 0.9809 0.9767 0.974 0.9789 0.9688 0.9826 0.9517 0.8551 0.0000 0.9827 0.9796 0.0041
± 0.0031 ± 0.0017 ± 0.0036 ± 0.0014 ± 0.0025 ± 0.0014 ± 0.0070 ± 0.0020 ± 0.0121 ± 0.0052 ± 0.0000 ± 0.0031 ± 0.0014 ± 0.0129
Yeast 0.0236 0.0000 0.0000 0.0193 0.0000 0.0000 0.0000 0.0513 0.0000 0.0000 0.0000 0.1649 0.0937 0.155
± 0.0499 ± 0.0000 ± 0.0000 ± 0.0408 ± 0.0000 ± 0.0000 ± 0.0000 ± 0.0543 ± 0.0000 ± 0.0000 ± 0.0000 ± 0.0957 ± 0.0929 ± 0.0766
Page blocks 0.8275 0.8142 0.8593 0.8302 0.1030 0.8411 0.8290 0.8180 0.6992 0.0473 0.0000 0.3131 0.7762 0.0000
± 0.0119 ± 0.0131 ± 0.0080 ± 0.0093 ± 0.0851 ± 0.0173 ± 0.0118 ± 0.0163 ± 0.0351 ± 0.0035 ± 0.0000 ± 0.0283 ± 0.0095 ± 0.0000
Shuttle 0.9195 0.9422 0.8999 0.9707 0.0028 0.9446 0.3690 0.9881 0.8915 0.0000 0.9400 0.9039 0.9205 0.9432
± 0.0766 ± 0.0826 ± 0.0931 ± 0.0165 ± 0.0064 ± 0.0805 ± 0.0521 ± 0.0088 ± 0.0080 ± 0.0000 ± 0.0577 ± 0.0871 ± 0.0726 ± 0.0591
Average 0.6691 0.6857 0.6911 0.6944 0.5533 0.6605 0.5896 0.6894 0.6550 0.5241 0.5167 0.6318 0.6596 0.5773
  1. The best performance is shown in italic for each dataset