Skip to main content

Table 2 Performance of gene feature selection methods with KNN classifier (high) and SVM classifier (low) in two-class datasets

From: Feature selection of gene expression data for Cancer classification using double RBF-kernels

Dataset: Gastric cancer
  DKBCGS GINI Χ2-Statistic Info.Gain KW RF MRMR KBCGS
ACC 0.9821 0.9664 0.9875 0.9779 0.9038 0.9548 0.9986 0.9716
TNR 1.0000 0.9500 0.9367 1.0000 0.9500 0.9800 1.0000 0.9755
TPR 0.9818 0.9677 0.9969 0.9759 0.8771 0.9498 1.0000 0.9826
TIME(s) 0.0846 0.7349 1.4736 0.7542 9.7452 4.2604 0.9007 0.6518
Dataset: DLBCL
  DKBCGS GINI Χ2-Statistic Info.Gain KW RF MRMR KBCGS
ACC 0.9833 0.9615 0.9865 0.9712 0.9123 0.9245 0.9341 0.9795
TNR 0.9943 0.9456 0.9422 0.9854 0.9457 0.9456 0.9654 1.0000
TPR 0.9863 0.9513 0.9645 0.9541 0.9024 0.9234 0.9432 0.9712
TIME(s) 0.1215 0.2257 0.1954 0.1857 0.1678 0.5111 0.0931 0.2148
Dataset: Gastric cancer
  DKBCGS GINI Χ2-Statistic Info.Gain KW RF MRMR KBCGS
ACC 1.0000 0.9768 0.9855 0.9623 0.9168 0.973 0.9988 0.9822
TNR 1.0000 0.9611 0.95 0.9158 0.9316 0.9433 1.0000 1.0000
TPR 1.0000 0.9929 0.9971 0.9776 0.9121 0.9827 1.0000 0.9755
TIME(s) 0.0846 0.7349 1.4736 0.7542 9.7452 4.2604 0.9007 0.7418
Dataset: DLBCL
  DKBCGS GINI Χ2-Statistic Info.Gain KW RF MRMR KBCGS
ACC 1.0000 1 0.9975 1.0000 0.9975 0.9750 0.9975 0.9845
TNR 1.0000 1.0000 1.0000 1.0000 0.9683 0.9733 0.9571 0.9579
TPR 1.0000 1.0000 1.0000 1.0000 0.8383 0.9437 0.9917 0.9931
TIME(s) 0.1215 0.2257 0.1954 0.1857 1.6478 0.5111 0.0931 0.2148