Skip to main content

Table 9 Top 20 important features extracted from female patients

From: Identification of clinical factors related to prediction of alcohol use disorder from electronic health records using feature selection methods

No

Feature

Description

Ranking

1

Age

Age at the relay study

0.481

2

LOS

Length of stay at the hospital

0.094

3

DK70

Alcoholic liver disease

0.037

4

DK74

Hepatic fibrosis

0.021

5

DA41

Other sepsis

0.020

6

ED

Visitor of Emergency Department

0.016

7

DZ03

Observation for suspected tuberculosis

0.014

8

Outpatient

Admitted to the hospital

0.014

9

DI63

Cerebral infarction due to thrombosis of precerebral arteries

0.013

10

Admission

Ambulatory care

0.012

11

DS82

Fracture of lower leg, including ankle

0.010

12

DI69

Sequelae of cerebrovascular disease

0.010

13

DK90

Intestinal malabsorption

0.010

14

DS53

Dislocation, sprain and strain of joints and ligaments of elbow

0.008

15

DK76

Other diseases of liver

0.007

16

DM19

Another arthrosis

0.007

17

DM24

Loose body in joint

0.007

18

DK50

Crohn disease [regional enteritis]

0.007

19

DK21

Gastro-oesophageal reflux disease

0.007

20

DS62

Fracture of first metacarpal bone

0.007