CPC G06Q 30/0185 (2013.01) [G06F 11/302 (2013.01); G06F 11/3438 (2013.01); G06F 21/316 (2013.01); G06F 21/554 (2013.01); G06Q 10/10 (2013.01); G06Q 20/027 (2013.01); G06Q 20/40145 (2013.01); G06Q 20/4016 (2013.01); G06Q 20/4093 (2013.01); G06Q 20/4097 (2013.01); G06Q 40/03 (2023.01); G06Q 50/265 (2013.01); H04L 63/1425 (2013.01); H04L 63/1433 (2013.01); H04L 67/535 (2022.05)] | 17 Claims |
1. A fraud prevention system for determining a fraudulent actor, the system comprising:
a fraud prevention server including an electronic processor and a memory, the memory including an online application origination (OAO) service,
wherein the OAO service is instructions that, when executed by the electronic processor, causes the electronic processor to perform a set of operations including
determining whether a first website configuration of a first online application has a multi-webpage flow;
responsive to determining that the first website configuration of the first online application does not have the multi-webpage flow, the set of operations further includes
calculating first input data on a single webpage of the first online application;
storing the first input data in the memory;
determining a first fraud risk score of the first online application based on the first input data and a first online application origination (OAO) model differentiating between a behavior of a first normal user and a behavior of a first actor during a submission of the first online application on a first device;
determining whether a second website configuration of a second online application has the multi-webpage flow;
responsive to determining that the second website configuration of the second online application has the multi-webpage flow, the set of operations further includes
calculating second input data on each webpage of a plurality of webpages of the second online application;
storing the second input data in the memory;
combining all of the second input data associated with a single session of the multi-webpage flow into multi-webpage input data;
storing the multi-webpage input data in the memory;
determining a second fraud risk score of the second online application based on the multi-webpage input data and a second online application origination (OAO) model differentiating between a behavior of a second normal user and a behavior of a second actor during a submission of the second online application on a second device; and
generating an approval signal, a hold signal, or a denial signal for the first online application or the second online application, based on the first fraud risk score or the second fraud risk score, respectively; and
transmitting the approval signal, the hold signal, or the denial signal for the first online application or the second online application to a client server.
|