Table 8 The overall results that are macro- and weighted-averaged, as well as the macro-averaged analysis of the results for each class of information provided in terms of Precision, Recall, and F1-measure.

From: Heart disease risk factors detection from electronic health records using advanced NLP and deep learning techniques

Risk factor

Indicator

Precision

Recall

F1-score

Support

Diabetes

a1c

0.67

0.92

0.77

64

Glucose

1.00

0.07

0.13

29

Mention

0.98

1.00

0.99

489

CAD

Event

0.71

0.76

0.74

173

Mention

0.84

0.93

0.88

183

Symptom

0.89

0.80

0.84

65

Test

0.89

0.12

0.21

25

Hypertension

High_bp

0.98

0.99

0.98

186

mention

1.00

0.99

0.99

478

Hyperlipidemia

High_chol

0.00

0.00

0.00

7

High_ldl

0.85

0.74

0.79

24

Mention

0.98

1.00

0.99

200

OBESE

Obese_bmi

0.00

0.00

0.00

9

Mention

0.93

1.00

0.97

107

Smoker

Smoker_current

0.00

0.00

0.00

36

Smoker_ever

0.00

0.00

0.00

3

Smoker_never

0.93

0.96

0.94

111

Smoker_past

0.78

0.85

0.81

110

Smoker_unknown

0.99

0.97

0.98

197

Medication

 

0.82

0.51

0.63

2062

Family history

 

0.87

0.90

0.88

13

Accuracy

   

0.9366

42,946

Macro average

 

0.3383

0.2920

0.2899

42,946

Weighted avg

 

0.9265

0.9366

0.9290

42,946

Micro average

   

0.9366

42,946