Skip to main content

Table 5 The minimum error rates yielded by k Nearest Neighbor classifier with feature selection methods along‐with the classification error without selection

From: A feature selection method for classification within functional genomics experiments based on the proportional overlapping score

Dataset

ISIS

Wil‐RS

mRMR

MP

POS

Full set

Leukaemia

0.064 (1)

0.074 (6)

0.135 (50)

0.019 (1)

0.005 (1)

0.109

Breast

0.385 (4)

0.405 (11)

0.404 (50)

0.346 (19)

0.332 (11)

0.405

Srbct

0.105 (2.63)

0.157 (3)

0.098 (48)

0.005 (26)

0.005 (22)

0.034

Lung

0.030 (3.26)

0.203 (12)

0.027 (49)

0.017 (17)

0.011 (12)

0.0005

GSE24514

0.074 (1.96)

0.032 (20)

0.041 (50)

0.036 (50)

0.039 (50)

0.041

GSE4045

0.239 (1.47)

0.066 (43)

0.207 (38)

0.137 (50)

0.142 (3)

0.103

GSE14333

0.425 (9)

0.420 (8)

0.455 (23)

0.450 (34)

0.438

GSE27854

0.432 (5)

0.420 (3)

0.454 (13)

0.420 (6)

0.464

  1. The numbers in brackets represent the size, average size for ISIS method, of the gene set that corresponding to the minimum error rate. Boldface numbers indicate the lowest error rate (the highest accuracy) among the compared methods for the corresponding datasets.