Skip to main content

Table 2 Comparison of overall accuracy on RNA-Seq gene expression datasets

From: BCDForest: a boosting cascade deep forest model towards the classification of cancer subtypes based on gene expression data

Code Dataset Sample Gene Class Overall Accuracy
KNN LR RF SVM gcForest BCDForest
1 PANCANCER 3594 8026 11 0.955 0.979 0.960 0.968 0.965 0.973
2 BRCA 514 3641 4 0.778 0.854 0.845 0.793 0.881 0.920
3 GBM 164 3180 4 0.694 0.651 0.702 0.619 0.741 0.806
4 LUNG 275 4000 3 0.710 0.744 0.791 0.786 0.830 0.867
5 COAD_I 264 3010 6 0.348 0.287 0.377 0.372 0.392 0.411
6 COAD_N 270 3006 3 0.699 0.631 0.696 0.700 0.711 0.730
7 COAD_T 282 3014 3 0.766 0.701 0.767 0.765 0.767 0.785
8 LIHC_I 347 4401 3 0.532 0.491 0.536 0.527 0.558 0.588
9 LIHC_N 400 4398 2 0.695 0.519 0.698 0.696 0.708 0.759
10 LIHC_T 347 4347 3 0.574 0.503 0.579 0.561 0.608 0.652