Research Article
Research on Credit Risk Prediction under Unbalanced Dataset Based on Ensemble Learning
Table 1
Features and description.
| | Feature | Type | Description |
| | Annual_inc | Numerical type | Annual income | | Dti | Numerical type | The ratio of the total debt paid by the borrower per month to the total debt | | Delinq_2 yrs | Numerical type | The number of borrowers whose credit files are more than 30 days overdue in the past two years | | Bc_open_to_buy | Numerical type | Permissible purchase quota of all bank cards | | Bc_util | Numerical type | Ratio of total current amount of all bank card accounts to credit limit | | Int_rate | Numerical type | Borrowing interest rate | | Total_bc_limit | Numerical type | Ratio of total credit limit to credit limit of bank card | | Open_acc | Numerical type | The number of open credit lines in the borrower’s credit file | | Pub_rec | Numerical type | Negative public record of the borrower | | Total_acc | Numerical type | The total number of credit lines in the borrower’s credit file to date | | Revol_bal | Numerical type | Total credit turnover balance | | Installment_feat | Numerical type | The ratio of the applicant’s total monthly repayment expenditure to the monthly income | | Home_ownership | Character type | House ownership status | | Verification_status | Character type | Whether the source of income is verified | | Desc | Character type | Description of loan | | Purpose | Character type | Purpose of borrowing | | Addr_state | Character type | Borrower’s location | | Pub_rec_bankruptcies | Character type | Number of public bankruptcy records |
|
|
