Skip to main content

Table 1 Artificial partial mislabeling introduced to the simulated data set to check robustness of FE

From: Principal component analysis-based unsupervised feature extraction applied to in silico drug discovery for posttraumatic stress disorder-mediated heart disease

 

(a)

 

(b)

 

(c)

Corr. coef.

0.92

 

0.68

 

0.60

Class

1

2

3

4

 

1

2

3

4

 

1

2

3

4

1

4

1

0

0

 

4

0

1

0

 

4

0

0

1

2

1

4

0

0

 

0

4

0

1

 

0

4

1

0

3

0

0

4

1

 

1

0

4

0

 

0

1

4

0

4

0

0

1

4

 

0

1

0

4

 

1

0

0

4

  1. Rows and columns represent true and modified labels. Numbers represent the number of samples with correct and modified labels. (a) little; (b) medium; and (c) heavy mislabeling. Correlation coefficients represent the amount of mislabeling (larger correlation coefficients correspond to less mislabeling).
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us