Risk-based machine learning classifier
The present disclosure relates generally to a risk-based fraud identification and risk analysis system. For example, the system may receive application data from a first borrower user, determine a segment associated with the application data, apply application data to one or more machine learning (ML) models, and receive a score based at least in part upon output of the ML model.
1. A method for computing an application score for a loan, the method comprising:
receiving, by a computer system, application data identifying a plurality of attributes of: a first borrower; a dealer; and a lender;
generating, by the computer system, a plurality of input features based upon the application data and historical application data;
applying the plurality of input features to a trained machine learning (ML) model to generate an output;
determining an application score based on the output of the trained ML model, wherein determining the application score comprises adjusting the output of the ML model to generate an adjusted application score based on at least one feature identified in evaluation of at least the application data and the historical application data, wherein adjusting the output of the ML model comprises:
identifying other similar applications sharing a same dealer/lender relationship;
clustering the identified applications; and
adjusting the output of the ML model based on the clustered applications;
determining a reason for how at least one of the input features affects the application score; and
providing, by the computer system, the adjusted application score and the at least one reason for how the at least one of the input features affects the application score to a dealer user device or to a lender user device.
2. The method of claim 1 , wherein the application data is received from a borrower user device.
3. The method of claim 1 , wherein the application data is received from a borrower user device and from the dealer user device.
4. The method of claim 3 , wherein the historical application data comprises at least one of: historical application scores; historical averages of loan characteristics; historical averages of borrower characteristics.
5. The method of claim 1 , wherein the historical application data is extracted from at least one historical application identified based, at least in part, on the application data.
6. The method of claim 5 , further comprising weighting the historical application data.
7. The method of claim 6 , wherein the historical application data is weighted based on an age of the data.
8. The method of claim 7 , wherein the historical application data is weighted based on the age of the data such that more recent historical application data has a greater weight than less recent historical application data.
9. The method of claim 6 , wherein the historical application data is weighted according to a risk profile.
10. The method of claim 1 , further comprising determining one or more actions for the application data based at least in part on the application score and on the historical application data.
11. A system comprising:
one or more processors; and
a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to:
receive application data identifying a plurality of attributes of: a first borrower; a dealer; and a lender;
generate a plurality of input features based upon the application data and historical application data;
apply the plurality of input features to a trained machine learning (ML) model to generate an output;
determine an application score based on the output of the trained ML model, wherein determining the application score comprises adjusting the output of the ML model to generate an adjusted application score based on at least one feature identified in evaluation of at least the application data and the historical application data, wherein adjusting the output of the ML model comprises:
identifying other similar applications sharing a same dealer/lender relationship;
clustering the identified applications; and
adjusting the output of the ML model based on the clustered applications;
determine at least one reason for how at least one of the input features affects the application score; and
provide the adjusted application score and the at least one reason for how the at least one of the input features affected the application score.
12. The system of claim 11 , wherein the application data is received from a borrower user device.
13. The system of claim 11 , wherein the application data is received from a borrower user device and from the dealer user device.
14. The system of claim 13 , wherein the historical application data comprises at least one of: historical application scores; historical averages of loan characteristics; historical averages of borrower characteristics.
15. The system of claim 11 , wherein the historical application data is extracted from at least one historical application identified based, at least in part, on the application data.
16. The system of claim 15 , further comprising weighting the historical application data.
17. The system of claim 16 , wherein the historical application data is weighted based on an age of the data.
18. The system of claim 17 , wherein the historical application data is weighted based on the age of the data such that more recent historical application data has a greater weight than less recent historical application data.
19. The system of claim 16 , wherein the historical application data is weighted according to a risk profile.
20. The system of claim 11 , further comprising determining one or more actions for the application data based at least in part on the application score and on the historical application data.