Skip to main content

Table 1 Threshold analysis for each record linkage tool

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

Method*

Threshold (TH)

Pairs above TH

Sensitivity

Specificity

FPs above TH

FNs below TH(%)

PPV

CIDACS-RL

0.8827056

3026 (46.86)

99.87

99.94

2 (0.07)

4 (0.13)

99.93

AtyImo

8777

3005 (46.54)

98.91

99.39

21 (0.70)

33 (1.09)

99.30

RecLink

0.8075590

2243 (34.74)

73.75

99.71

10 (0.45)

795 (26.25)

99.55

Febrl

3722604

2832 (43.86)

90.58

97.40

89 (3.14)

285 (9.41)

96.86

FRILL

48

2351 (36.41)

74.66

97.36

90 (3.83)

767 (25.33)

96.17

  1. *Execution time (in minutes): CIDACS-RL < 1, AtyImo = 28, RecLink < 1, FRIL = 7, and Febrl = 130