Skip to main content

Table 1 The distribution of pairs for each corpus according to classification success level using cross-validation setting

From: A detailed error analysis of 13 kernel methods for protein-protein interaction extraction

 

AIMed

BioInfer

HPRD50

IEPA

LLL

 

Total

T

F

T, %

F, %

Total

T

F

T, %

F, %

Total

T

F

T, %

F, %

Total

T

F

T, %

F, %

Total

T

F

T, %

F, %

0

77

73

4

7.3%

0.1%

58

44

14

1.7%

0.2%

4

1

3

0.6%

1.1%

2

1

1

0.3%

0.2%

5

0

5

0.0%

3.0%

1

95

89

6

8.9%

0.1%

158

107

51

4.2%

0.7%

7

4

3

2.5%

1.1%

13

5

8

1.5%

1.7%

7

0

7

0.0%

4.2%

2

105

101

4

10.1%

0.1%

206

130

76

5.1%

1.1%

12

8

4

4.9%

1.5%

11

3

8

0.9%

1.7%

27

0

27

0.0%

16.3%

3

121

104

17

10.4%

0.4%

306

198

108

7.8%

1.5%

18

7

11

4.3%

4.1%

26

13

13

3.9%

2.7%

10

0

10

0.0%

6.0%

4

139

115

24

11.5%

0.5%

349

203

146

8.0%

2.0%

26

10

16

6.1%

5.9%

30

10

20

3.0%

4.1%

16

0

16

0.0%

9.6%

5

140

91

49

9.1%

1.0%

440

225

215

8.9%

3.0%

20

12

8

7.4%

3.0%

43

19

24

5.7%

5.0%

21

2

19

1.2%

11.4%

6

142

70

72

7.0%

1.5%

481

209

272

8.2%

3.8%

33

9

24

5.5%

8.9%

61

22

39

6.6%

8.1%

26

1

25

0.6%

15.1%

7

176

65

111

6.5%

2.3%

619

248

371

9.8%

5.2%

35

15

20

9.2%

7.4%

51

20

31

6.0%

6.4%

29

8

21

4.9%

12.7%

8

248

72

176

7.2%

3.6%

785

256

529

10.1%

7.4%

37

9

28

5.5%

10.4%

79

31

48

9.3%

10.0%

19

6

13

3.7%

7.8%

9

372

69

303

6.9%

6.3%

876

245

631

9.7%

8.8%

46

10

36

6.1%

13.3%

99

32

67

9.6%

13.9%

26

15

11

9.1%

6.6%

10

461

47

414

4.7%

8.6%

1067

204

863

8.1%

12.1%

61

33

28

20.2%

10.4%

101

38

63

11.3%

13.1%

31

19

12

11.6%

7.2%

11

619

29

590

2.9%

12.2%

1061

164

897

6.5%

12.6%

49

19

30

11.7%

11.1%

112

46

66

13.7%

13.7%

32

32

0

19.5%

0.0%

12

1002

43

959

4.3%

19.8%

1390

183

1207

7.2%

16.9%

57

13

44

8.0%

16.3%

106

47

59

14.0%

12.2%

45

45

0

27.4%

0.0%

13

2137

32

2105

3.2%

43.5%

1870

118

1752

4.7%

24.6%

28

13

15

8.0%

5.6%

83

48

35

14.3%

7.3%

36

36

0

22.0%

0.0%

  1. The distribution of pairs (total, positive and negative) in terms of the number of kernels that classify them correctly. Results shown for each corpus separately. Aggregated results are shown in Figure 1. All the 13 kernels are taken into consideration.