Systems and methods for optimizing transaction conversion rate using machine learning

Abstract

A method for optimizing transaction authorization conversion rates using machine learning includes retrieving payment transaction parameters and authorization results for a plurality of past payment transactions from a database, generating a transaction success model comprising authorization success factors for each of a plurality of payment transaction parameters using a machine learning training phase based on the retrieved payment transaction parameters and authorization results, receiving, at an acquirer processor, a payment transaction from a merchant, modifying one or more parameters of the payment transaction according to the generated transaction success model, and submitting the modified payment transaction to a financial institution for processing.

1 - 20 . (canceled) 
     
     
         21 . A computer-implemented method for optimizing authorization transaction conversion rates, comprising:
 receiving a request for authorization of a transaction from a user device;   determining patterns of acceptance or denial of the request based, at least in part, on processing of transaction parameters and authorization results for a plurality of past transactions from a dataset;   inputting the determined patterns into a machine learning model to generate a transaction success model, wherein the transaction success model includes authorization success factors for a plurality of transactions;   applying, by the machine learning model, the transaction success model to the request and re-format one or more parameters associated with the request; and   transmitting the re-formatted request to a service provider for authorization of the transaction.   
     
     
         22 . The computer-implemented method of  claim 21 , wherein generating the transaction success model, further comprises:
 processing, by the machine learning model, the plurality of past transactions in the dataset to tune the transaction success model;   determining, by the machine learning model, the transaction success model provides an improvement in the transaction conversion rates; and   applying, by the machine learning model, the transaction success model to the parameters of the request based, at least in part, on the determination.   
     
     
         23 . The computer-implemented method of  claim 21 , further comprising:
 determining an inclusion, an exclusion, or an alteration of the one or more parameters associated with the request results in an improvement or a decrease in rate of transaction authorization.   
     
     
         24 . The computer-implemented method of  claim 21 , further comprising:
 automatically calibrating optimization factors of the transaction success model through analysis of the authorization success factors, one or more transaction scenarios, and issuer negative results.   
     
     
         25 . The computer-implemented method of  claim 21 , further comprising:
 determining, by the machine learning model, transaction request with tokens, encrypted information, or a combination thereof improves probability of transaction authorization; and   re-formatting, by the machine learning model, the one or more parameters associated with the request to include a token, an encrypted authorization credentials, or a combination thereof.   
     
     
         26 . The computer-implemented method of  claim 21 , further comprising:
 determining, by the machine learning model, at least one network with a higher probability of transaction authorization; and   selecting, by the machine learning model, the at least one network for transmitting the re-formatted request to the service provider for transaction authorization.   
     
     
         27 . The computer-implemented method of  claim 21 , further comprising:
 receiving a transaction authorization result for the re-formatted request from the service provider; and   adding the transaction authorization result and the parameters of the re-formatted request to the dataset, wherein the transaction authorization result and the parameters are utilized for subsequent transactions.   
     
     
         28 . The computer-implemented method of  claim 21 , wherein the one or more parameters associated with the request are re-formatted in batches, in a queue of the requests, in real-time, or asynchronously. 
     
     
         29 . The computer-implemented method of  claim 21 , wherein the dataset includes an authorization result that indicates a specific combination of the transaction parameters that resulted in authorization of the request and a reason response code. 
     
     
         30 . The computer-implemented method of  claim 21 , wherein the transaction parameters include one or more of a billing address, a card verification value (CVV), a payment processing network, a payment vehicle expiration date, a payment vehicle issuer token, and a merchant classification code (MCC). 
     
     
         31 . A system for optimizing authorization transaction conversion rates, comprising:
 one or more processors; and   at least one non-transitory computer readable medium storing instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
 receiving a request for authorization of a transaction from a user device; 
 determining patterns of acceptance or denial of the request based, at least in part, on processing of transaction parameters and authorization results for a plurality of past transactions from a dataset; 
 inputting the determined patterns into a machine learning model to generate a transaction success model, wherein the transaction success model includes authorization success factors for a plurality of transactions; 
 applying, by the machine learning model, the transaction success model to the request and re-format one or more parameters associated with the request; and 
 transmitting the re-formatted request to a service provider for authorization of the transaction. 
   
     
     
         32 . The system of  claim 31 , wherein generating the transaction success model, further comprises:
 processing, by the machine learning model, the plurality of past transactions in the dataset to tune the transaction success model;   determining, by the machine learning model, the transaction success model provides an improvement in the transaction conversion rates; and   applying, by the machine learning model, the transaction success model to the parameters of the request based, at least in part, on the determination.   
     
     
         33 . The system of  claim 31 , further comprising:
 determining an inclusion, an exclusion, or an alteration of the one or more parameters associated with the request results in an improvement or a decrease in rate of transaction authorization.   
     
     
         34 . The system of  claim 31 , further comprising:
 automatically calibrating optimization factors of the transaction success model through analysis of the authorization success factors, one or more transaction scenarios, and issuer negative results.   
     
     
         35 . The system of  claim 31 , further comprising:
 determining, by the machine learning model, transaction request with tokens, encrypted information, or a combination thereof improves probability of transaction authorization; and   re-formatting, by the machine learning model, the one or more parameters associated with the request to include a token, an encrypted authorization credentials, or a combination thereof.   
     
     
         36 . The system of  claim 31 , further comprising:
 determining, by the machine learning model, at least one network with a higher probability of transaction authorization; and   selecting, by the machine learning model, the at least one network for transmitting the re-formatted request to the service provider for transaction authorization.   
     
     
         37 . The system of  claim 31 , further comprising:
 receiving a transaction authorization result for the re-formatted request from the service provider; and   adding the transaction authorization result and the parameters of the re-formatted request to the dataset, wherein the transaction authorization result and the parameters are utilized for subsequent transactions.   
     
     
         38 . A non-transitory computer readable medium for optimizing authorization transaction conversion rates, the non-transitory computer readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform operations comprising:
 receiving a request for authorization of a transaction from a user device;   determining patterns of acceptance or denial of the request based, at least in part, on processing of transaction parameters and authorization results for a plurality of past transactions from a dataset;   inputting the determined patterns into a machine learning model to generate a transaction success model, wherein the transaction success model includes authorization success factors for a plurality of transactions;   applying, by the machine learning model, the transaction success model to the request and re-format one or more parameters associated with the request; and   transmitting the re-formatted request to a service provider for authorization of the transaction.   
     
     
         39 . The non-transitory computer readable medium of  claim 38 , wherein generating the transaction success model, further comprises:
 processing, by the machine learning model, the plurality of past transactions in the dataset to tune the transaction success model;   determining, by the machine learning model, the transaction success model provides an improvement in the transaction conversion rates; and   applying, by the machine learning model, the transaction success model to the parameters of the request based, at least in part, on the determination.   
     
     
         40 . The non-transitory computer readable medium of  claim 38 , further comprising:
 determining an inclusion, an exclusion, or an alteration of the one or more parameters associated with the request results in an improvement or a decrease in rate of transaction authorization.