Credit-Risk Prediction Model Using Hybrid Deep—Machine-Learning Based Algorithms

Table 1

Inputs and their representation.

Inputs	Representation of inputs
Age	Age of customer
Gender	Gender of customer
Marital status	Customer’s marital status: married: 0 and not married: 1, others: 2
Job	Job of customers
Education	Diploma: 0, degree holder: 1,second degree: 2,Ph.D.: 3
work experience	<5, 5–10, 11–15, 16–20, 21–25 and >25
Type of credit	Mortgage,auto,personal,education,agriculture,medical
Annual income	Annual income of the customers
Coapplicant income	The annual income of the customers coapplicant
Total income	The sum of annual income and the coapplicant annual income
Duration of credit	0–60, 61–120, 121–180, 181–240, and 241–300
Collateral	Physical asset: 0, salary: 1
Credit history	0, 1, 2, 3, 4, and 5
Credit grade	A : 0, B : 1, C : 2, and D : 3
Amount of credit	The amount of credit based on the total income of the customers
Credit interest rate	13%, 11%, and 7%