Skip to main content

Table 2 Case study (CadUnico x SINAN-TB) dataset: linkage analysis

From: CIDACS-RL: a novel indexing search and scoring-based record linkage system for huge datasets with high accuracy and scalability

Cut-off

Specificity

Sensitivity

Matches (%)

True matches (%)

False matches (%)

Missed true matches (%)

0.860

75.0

97.1

16,443 (55.15)

12,100 (73.59)

4343 (26.41)

361 (2.90)

0.870

82.2

95.5

14,984 (50.25)

11,901 (79.42)

3083 (20.58)

560 (4.49)

0.880

87.7

94.5

13,901 (46.62)

11,770 (84.67)

2131 (15.33)

691 (5.55)

0.890

91.8

93.3

13,046 (43.76)

11,621 (89.08)

1425 (10.92)

840 (6.74)

0.896

93.5

92.5

12,661 (42.46)

11,532 (91.08)

1129 (8.92)

929 (7.46)

0.900

94.2

91.7

12,423 (41.67)

11,424 (91.96)

999 (8.04)

1037 (8.32)

0.910

95.8

89.8

11,931 (40.02)

11,194 (93.82)

737 (6.18)

1267 (10.17)

0.920

96.7

88.1

11,546 (38.72)

10,972 (95.03)

574 (4.97)

1489 (11.95)

0.930

98.0

85.4

10,984 (36.84)

10,636 (96.83)

348 (3.17)

1825 (14.65)