Skip to main content

Table 1 Number of instances for each class before and after applying SMOTE technique

From: HormoNet: a deep learning approach for hormone-drug interaction prediction

 

Before SMOTE

After SMOTE

Samples

Class A

Class B

Class C

Samples

Class A

Class B

Class C

All

4773

561

3701

511

11,103

3701

3701

3701

Train

3579

429

2778

372

8327

2785

2765

2777

Test

1194

132

923

139

2776

916

936

924

  1. The number of samples in each classes in the data set is not the same, therefore this SMOTE balanced the number of samples in all classes are highlighted in italic
  2. For risk level of interaction of drug A and drug B: class A is major, class B is moderate, and class C is minor
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us