Skip to main content

Table 6 Summary of the experimental results of original dataset and datasets with different percentages of missing data-values are average of the corresponding values in Table 5

From: Semantics-based plausible reasoning to extend the knowledge coverage of medical knowledge bases for improved clinical decision support

Reasoning Strategy

Deductive Only

Deductive + Plausible (Without Expert’s Verification)

Deductive + Plausible (With Expert’s Verification)

Deductive Only

Deductive + Plausible (Without Expert’s Verification)

Deductive + Plausible (With Expert’s Verification)

Deductive Only

Deductive + Plausible (Without Expert’s Verification)

Deductive + Plausible (With Expert’s Verification)

Deductive Only

Deductive + Plausible (Without Expert’s Verification)

Deductive + Plausible (With Expert’s Verification)

Deductive Only

Deductive + Plausible (Without Expert’s Verification)

Deductive + Plausible (With Expert’s Verification)

 

Original Dataset

5% Missing

10% Missing

15% Missing

20% Missing

#Queries answered correctly (Avg.)

144

149

147

140

149

143

133

152

143

122

149

142

113

148

135

#Queries answered incorrectly (Avg.)

7

5

6

8

4

8

7

2

7

9

5

9

10

6

11

#Queries unanswered (Avg.)

4

1

2

8

1

4

15

1

5

24

1

5

32

0

8

# Plausible Rules (Avg.)

N/A

3

2

N/A

4

2

N/A

5

2

N/A

6

4

N/A

7

4

# Plausible Facts (Avg.)

N/A

1

1

N/A

2

2

N/A

2

4

N/A

2

2

N/A

3

4

Coverage (Avg.)

97%

99%

99%

95%

99%

97%

90%

99%

97%

85%

99%

97%

79%

100%

95%

Correctness (Avg.)

95%

97%

96%

95%

97%

95%

95%

98%

95%

93%

97%

94%

92%

96%

92%