Improved ML-Based Technique for Credit Card Scoring in Internet Financial Risk Control

Table 6

Variables selected into the model.

Bank of deposit	0.053
Number of credit cards	0.052
Maximum credit card limit in recent 1 month	0.084
Maximum overdue days of short-term loans	0.286
The salary per month	0.249
The standard deviation of the number of SMS messages sent at night in the last three months	0.072
The standard deviation of the frequency of answering unlabeled numbers at night in recent two months	0.075
Debit card ratio	0.073
Bill number	0.069
Amount to be paid under credit products	0.065
Average consumption in recent 30 days	0.063
Total data months	0.068
The proportion of credit cards with bills in the last 60 days	0.066
Balance of credit products	0.062
Percentage standard deviation of dialing all numbers at night in recent 60 days	0.060
Bank of deposit	0.061