Skip to main content

Table 9 Detailed results obtained by the Rule-based system

From: Automatic de-identification of French electronic health records: a cost-effective approach exploiting distant supervision and deep learning models

Class

Precision

Recall

F1-Score

Support

DATE

0.9927

0.9972

0.9950

2178

DOCTOR

0.9535

0.9196

0.9362

1293

VILLE

0.9713

0.9837

0.9775

860

PHONE

0.9967

0.9869

0.9918

612

PATIENT

0.5530

0.8375

0.6661

480

ZIP

0.9968

0.9231

0.9585

338

STR

0.9105

0.6431

0.7538

269

EMAIL

0.9135

1.0000

0.9548

95

micro avg

0.9275

0.9458

0.9365

6125

macro avg

0.9110

0.9114

0.9042

6125

weighted avg

0.9427

0.9458

0.9408

6125

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us