Method for improving operating profits with better automated decision making with artificial intelligence

Abstract

A business method to reverse an authorization request denial made according to general guidelines if the particular customer and the particular transaction pass various threshold tests. Alternatively, each customer is assigned different and independent personal thresholds for transaction types, amounts, times, locations, and context. These thresholds are then applied if an instant payment transaction request is about to be declined.

1 . A method for deciding if an adverse fraud score for a particular customer in a particular transaction should be overruled given that particular customer's business value and the consequential impact to the company making a decision to authorize that transaction, comprising:
 keeping available transaction record attributes of past business transactions with a plurality of customers for fast access with a network server using an apparatus for executing an algorithm for creating, initiating, and updating a plurality of corresponding and matching smart agents each having a non-transitory profile and individualized thresholds in a computer storage memory;   receiving a transaction record and identifying a particular customer from attributes in the record with the apparatus for executing an algorithm for creating, initiating, and updating a plurality of corresponding and matching smart agents;   determining a fraud score for a transaction risk represented in the transaction record with an apparatus for executing an algorithm for finding and matching a profile in computer memory storage for one of the plurality of corresponding and matching smart agents   determining a profitability score for a business value represented in attributes of the transaction record with an apparatus for executing an algorithm for using the one smart agent to set a comparison threshold stored in computer memory storage for each of a customer value, an impact, and a projected revenue loss;   determining between outputting a decision to approve the transaction, or deciding to decline the transaction, using an apparatus for executing an algorithm that tests the comparison thresholds stored in computer memory storage for each customer value, impact, and projected revenue loss; and   outputting a decision through a financial network server.   
     
     
         2 . The Method of  claim 1 , further comprising:
 an apparatus for executing an algorithm to maintain a running average business total of every customer in a corresponding smart agent profile maintained in computer storage memory and for comparing a single such customer and smart agent profile to such running average business total to a first threshold data value also maintained in computer storage memory, and for allowing a transaction declined decision to issue as a result of a first comparison.   
     
     
         3 . The Method of  claim 2 , further comprising:
 an apparatus for executing an algorithm for comparing a second threshold data value maintained in computer storage memory to a transaction amount reported in the transaction record received, and for allowing a transaction approved decision to issue as a result of a second comparison.   
     
     
         4 . The Method of  claim 3 , further comprising:
 an apparatus for executing an algorithm to maintain a pattern of transaction types in a record for every customer in a corresponding smart agent profile maintained in computer storage memory and for later comparing a current transaction type from the received transaction record to a transaction pattern as a third data value of single such customer and smart agent profile, and for allowing a transaction declined decision to issue as a result of a third comparison.   
     
     
         5 . The Method of  claim 4 , further comprising:
 an apparatus for executing an algorithm for comparing a fourth threshold data value maintained in computer storage memory to a transaction amount reported in the transaction record received, and for allowing a transaction approved decision to issue as a result of a fourth comparison.   
     
     
         6 . The Method of  claim 5 , further comprising:
 an apparatus for executing an algorithm to maintain an inventory of characteristics of an accountholders user devices in a record for every customer in a corresponding smart agent profile maintained in computer storage memory, and for later comparing a user device characteristic employed in the current transaction from the received transaction record as a fifth data value to the characteristics of user devices of a single such customer and smart agent profile, and for allowing a transaction declined decision to issue as a result of a fifth comparison.   
     
     
         7 . The Method of  claim 6 , further comprising:
 an apparatus for executing an algorithm for comparing a sixth threshold data value maintained in computer storage memory to a transaction amount reported in the transaction record received, and for allowing a transaction approved decision to issue as a result of a sixth comparison.   
     
     
         8 . The Method of  claim 1 , further comprising:
 an apparatus for executing an algorithm to maintain a smart agent profile maintained in computer storage memory corresponding to particular customers who represent a better-than-average level of business revenue and profits, wherein a larger-than-average number of transaction records have been processed and contributed to such smart agent profiles and result in more accurate decision making.