Systems and methods for a context-driven electronic transactions fraud detection

Abstract

Systems and methods are disclosed for establishing a multi-dimensional fraud detection system and payment analysis. One method includes: receiving transaction history of a user, the transaction history including a first payment vehicle and a second payment vehicle; determining, of the received transaction history, one or more instances of switching from one the first payment vehicle to the second payment vehicle; and determining a user-specific abandonment score for the user, based on the determined instances of switching from the first payment vehicle to the second payment vehicle.

1 - 20 . (canceled) 
     
     
         21 . A computer-implemented method for training a fraud detection system to manage fraudulent transactions, the method comprising:
 receiving an authorization request for at least one online transaction, wherein the authorization request includes transaction data;   processing the transaction data to determine a fraud analysis profile for at least one user associated with the online transaction;   calculating a fraud risk score, an abandonment score, a risk tolerance score, or a combination thereof based, at least in part, on the transaction data and the fraud analysis profile; and   comparing the fraud risk score with the abandonment score and the risk tolerance score to determine to approve the authorization request if the fraud risk score is determined to be lower than the abandonment score and the risk tolerance score.   
     
     
         22 . The computer-implemented method of  claim 21 , further comprising:
 determining the fraud analysis profile for the at least one user associated with the online transaction is not available;   generating a unique hash and a fraud analysis profile request; and   transmitting the fraud analysis profile request including the unique hash to generate the fraud analysis profile for the at least one user associated with the online transaction is not available.   
     
     
         23 . The computer-implemented method of  claim 21 , wherein calculating the fraud risk score further comprising:
 determining contextual information associated with the at least one online transaction, wherein the contextual information includes device-specific information, transaction history information, or a combination thereof; and   calculating the fraud risk score based, at least in part, on a comparison between the determined contextual information and stored contextual information, wherein a high fraud risk score is assigned to the at least one user upon determining inconsistencies during the comparison.   
     
     
         24 . The computer-implemented method of  claim 21 , wherein calculating the abandonment score further comprising:
 processing transaction history of the at least one user associated with the online transaction to determine a switching pattern of payment vehicles upon denial of the at least one online transaction; and   calculating the abandonment score based, at least in part, on the switching pattern, wherein a low abandonment score is assigned to the at least one user upon determining a higher switching pattern of the payment vehicles.   
     
     
         25 . The computer-implemented method of  claim 24 , further comprising:
 detecting an instance of switching between the payment vehicles by the at least one user during the online transaction; and   reducing the calculated abandonment score based, at least in part, on the detection, wherein the reduction of the calculated abandonment score is predetermined and is relative to a number of switching between the payment vehicles.   
     
     
         26 . The computer-implemented method of  claim 24 , further comprising:
 determining billing information associated with the payment vehicles; and   increasing the fraud risk score upon determining a discrepancy in the billing information between the payment vehicles.   
     
     
         27 . The computer-implemented method of  claim 21 , wherein calculating the risk tolerance score further comprising:
 determining transaction history of the at least one user, wherein the transaction history includes fraudulent activities associated with the at least one user; and   calculating the risk tolerance score based, at least in part, on the determined transaction history, preference information of a service provider, or a combination thereof, wherein a low risk tolerance score is assigned to the at least one user upon determining at least one incidence of fraudulent activity in the transaction history.   
     
     
         28 . The computer-implemented method of  claim 27 , wherein the risk tolerance score is adjusted at a predetermined time interval. 
     
     
         29 . The computer-implemented method of  claim 21 , wherein generating the fraud analysis profile for the at least one user further comprising:
 aggregating a plurality of transaction data associated with a plurality of authorization requests, a plurality of fraudulent activities, or a combination thereof associated with the at least one user; and   generating the fraud analysis profile based, at least in part, on a processing of the plurality of transaction data, the plurality of fraudulent activities, or a combination thereof.   
     
     
         30 . The computer-implemented method of  claim 29 , further comprising:
 tokenizing account identifying information associated with the aggregated plurality of transaction data;   transmitting the tokenized data for analysis to generate the fraud analysis profile; and   de-tokenizing the generated fraud analysis profile for storing in a profile database.   
     
     
         31 . A decentralized computer system for training a fraud detection system to manage fraudulent transactions, the method comprising:
 a data storage device storing instructions for training the fraud detection system to manage fraudulent transactions; and   a processor configured to execute the instructions to perform a method including:
 receiving an authorization request for at least one online transaction, wherein the authorization request includes transaction data; 
 processing the transaction data to determine a fraud analysis profile for at least one user associated with the online transaction; 
 calculating a fraud risk score, an abandonment score, a risk tolerance score, or a combination thereof based, at least in part, on the transaction data and the fraud analysis profile; and 
 comparing the fraud risk score with the abandonment score and the risk tolerance score to determine to approve the authorization request if the fraud risk score is determined to be lower than the abandonment score and the risk tolerance score. 
   
     
     
         32 . The system of  claim 31 , further comprising:
 determining the fraud analysis profile for the at least one user associated with the online transaction is not available;   generating a unique hash and a fraud analysis profile request; and   transmitting the fraud analysis profile request including the unique hash to generate the fraud analysis profile for the at least one user associated with the online transaction is not available.   
     
     
         33 . The system of  claim 31 , wherein calculating the fraud risk score further comprising:
 determining contextual information associated with the at least one online transaction, wherein the contextual information includes device-specific information, transaction history information, or a combination thereof; and   calculating the fraud risk score based, at least in part, on a comparison between the determined contextual information and stored contextual information, wherein a high fraud risk score is assigned to the at least one user upon determining inconsistencies during the comparison.   
     
     
         34 . The system of  claim 31 , wherein calculating the abandonment score further comprising:
 processing transaction history of the at least one user associated with the online transaction to determine a switching pattern of payment vehicles upon denial of the at least one online transaction; and   calculating the abandonment score based, at least in part, on the switching pattern, wherein a low abandonment score is assigned to the at least one user upon determining a higher switching pattern of the payment vehicles.   
     
     
         35 . The system of  claim 34 , further comprising:
 detecting an instance of switching between the payment vehicles by the at least one user during the online transaction; and   reducing the calculated abandonment score based, at least in part, on the detection, wherein the reduction of the calculated abandonment score is predetermined and is relative to a number of switching between the payment vehicles.   
     
     
         36 . The system of  claim 34 , further comprising:
 determining billing information associated with the payment vehicles; and   increasing the fraud risk score upon determining a discrepancy in the billing information between the payment vehicles.   
     
     
         37 . The system of  claim 31 , wherein calculating the risk tolerance score further comprising:
 determining transaction history of the at least one user, wherein the transaction history includes fraudulent activities associated with the at least one user; and   calculating the risk tolerance score based, at least in part, on the determined transaction history, preference information of a service provider, or a combination thereof, wherein a low risk tolerance score is assigned to the at least one user upon determining at least one incidence of fraudulent activity in the transaction history.   
     
     
         38 . A non-transitory machine-readable medium storing instructions that, when executed by a server, cause the server to perform a method for training a fraud detection system to manage fraudulent transactions, the method comprising:
 receiving an authorization request for at least one online transaction, wherein the authorization request includes transaction data;   processing the transaction data to determine a fraud analysis profile for at least one user associated with the online transaction;   calculating a fraud risk score, an abandonment score, a risk tolerance score, or a combination thereof based, at least in part, on the transaction data and the fraud analysis profile; and   comparing the fraud risk score with the abandonment score and the risk tolerance score to determine to approve the authorization request if the fraud risk score is determined to be lower than the abandonment score and the risk tolerance score.   
     
     
         39 . The non-transitory machine readable medium of  claim 38 , further comprising:
 determining the fraud analysis profile for the at least one user associated with the online transaction is not available;   generating a unique hash and a fraud analysis profile request; and   transmitting the fraud analysis profile request including the unique hash to generate the fraud analysis profile for the at least one user associated with the online transaction is not available.   
     
     
         40 . The non-transitory machine readable medium of  claim 38 , wherein calculating the fraud risk score further comprising:
 determining contextual information associated with the at least one online transaction, wherein the contextual information includes device-specific information, transaction history information, or a combination thereof; and   calculating the fraud risk score based, at least in part, on a comparison between the determined contextual information and stored contextual information, wherein a high fraud risk score is assigned to the at least one user upon determining inconsistencies during the comparison.