Skip to main content

Table 4 Experimental results

From: Construction and application of a knowledge graph-based question answering system for Nanjing Yunjin digital resources

Number

Intent name

Precision (%)

Recall (%)

F1

1

Ask the successor

92.59

98.04

0.952

2

Ask parts

96.55

96.55

0.966

3

Ask the formula

94.12

96.00

0.951

4

Ask the sorting method

94.20

94.20

0.942

5

Ask for definition

96.00

96.97

0.965

6

Ask for meaning

95.38

92.54

0.939

7

Ask the pattern

98.44

92.65

0.955

8

Ask the category of work

93.75

95.74

0.947

9

General evaluation data

95.33

95.28

0.953