Skip to main content

Table 1 Missing attribute values in the dataset

From: The 30-days hospital readmission risk in diabetic patients: predictive modeling with machine learning classifiers

Attribute

Type

Description

Missing rate%

Race

Nominal

Ethnicity, including Caucasian, Asian, African American, Hispanic, and others

2

Weight

Numeric

Weight (pounds)

97

Payer code

Nominal

Integer identifiers corresponding to 23 different values

52

Medical specialty

Nominal

Doctor professionals, such as internal medicine, surgery, and family doctors

53

Diagnosis 1

Nominal

Initial diagnosis (coded as the first three digits of ICD-9), a total of 848 different values

0.5

Diagnosis 2

Nominal

Secondary diagnosis (coded as the first three digits of ICD-9), a total of 923 different values

0.5

Diagnosis 3

Nominal

Additional secondary diagnosis (coded as the first three digits of ICD-9) for a total of 954 different values

1