Skip to main content

Table 4 Results of EXTRAE Algorithm on HUF corpus using different seed sizes, based on their F-Measure, AUC-ROC, and AU-PRC

From: Semi-supervised incremental learning with few examples for discovering medical association rules

HUF corpus

Seed size

Iterations

p Value

F-Measure

AUC-ROC

AU-PRC

5

3

4.79E-13

0.73

0.80

0.81

10

7

4.79E-13

0.75

0.80

0.81

15

8

3.67E-13

0.72

0.79

0.80

20

14

3.67E-13

0.73

0.80

0.81

25

8

5.34E-10

0.74

0.79

0.80

35

6

3.3E-6

0.73

0.78

0.80

50

13

3.3E-6

0.74

0.78

0.80

75

4

3.35E-9

0.72

0.79

0.81

100

5

3.35E-9

0.69

0.79

0.81

125

5

3.35E-9

0.74

0.79

0.80

150

5

3.67E-13

0.72

0.79

0.80

175

4

3.67E-13

0.72

0.79

0.80

200

6

3.67E-13

0.74

0.80

0.81

  1. Iterations is the max number of iterations reached and p value is obtained automatically using the filter approach on the seed set. Best results appear in boldface