Skip to main content

Table 4 The effectiveness of our approach in comparison with the simple Apriori under \(N = 8\) and \({min\_sup = 0}\)

From: Efficacy-specific herbal group detection from traditional Chinese medicine prescriptions via hierarchical attentive neural network model

Efficacy

The simple Apriori

Apriori on distilled prescriptions

\(K=2\) (%)

\(k=3\) (%)

\(k=4\) (%)

\(k=2\) (%)

\(k=3\) (%)

\(k=4\) (%)

Activating blood to

10.30

14.09

14.32

15.78±0.18

22.79±0.24

25.77±0.25

Resolve stasis

      

Dispersing phlegm

15.45

21.34

23.17

18.08±0.16

25.27±0.18

28.53±0.22

Strengthening bone

8.51

11.83

13.36

11.56±0.14

17.46±0.24

20.42±0.34

Stopping bleeding

18.62

23.58

24.56

19.13±0.27

23.75±0.42

25.27±0.49

Diminishing swelling

5.99

6.77

6.68

6.58±0.15

8.44±0.19

9.42±0.20

Draining dampness

17.98

22.25

22.63

19.48±0.17

23.99±0.20

24.90±0.19

Improving eyesight

10.37

13.67

15.00

13.86±0.18

19.37±0.20

21.52±0.21

Invigorating spleen

17.12

24.46

27.09

21.32±0.17

31.35±0.15

36.21±0.21

Arresting cough

15.81

20.85

22.01

19.78±0.16

27.52±0.26

31.09±0.35

Tranquillization

15.78

20.76

21.97

19.90±0.11

27.72±0.11

31.90±0.13

Relieving pain

8.88

12.27

14.30

9.88±0.20

12.92±0.28

14.30±0.31

Relieving itching

7.74

10.81

11.77

8.39±0.22

11.82±0.38

13.47±0.46

Removing obstruction

14.62

19.68

20.91

18.72±0.07

25.00±0.09

26.93±0.08

in the channels

      

The whole average

12.86

17.11

18.29

15.57

21.34

23.83