Skip to main content

Table 3 The number of original features and the number of features after dimension reduction

From: Improved cost-sensitive representation of data for solving the imbalanced big data classification problem

 

Name of data set

Main feature difference

Features after dimension reduction

1

Cancer

9

4

2

Wine

13

6

3

Ionosphere

34

17

4

Pima Indian diabetes

8

4

5

Iris

4

2

6

Wdbc

30

15

7

Cleveland

13

6

8

Musk

166

83

9

Dermatology-6

34

17

10

FuelCons

37

18

11

Movement_libras

90

45

12

Sonar

60

30

13

SPECTF

44

22

14

Colon tumor

166

83

15

DLBCL

5469

2734

16

Mnist

784

392

17

Caltech101

784

392

18

Kddcup-rootkit-imap-vs-back

41

20

19

Kddcup-buffer-overflow-vs-back

41

20

20

Kddcup-guess-passwd-vs-satan

41

20

21

Kddcup-land-vs-satan

41

20