Back to results
US 2020/0051087 A1Published application
Method for improving operating profits with better automated decision making with artificial inteligence
AssigneeBrighterion, Inc.
InventorAkli Adjaoute
Life of this patent
Estimating the term…
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.
Claims20
At least one computer network server for supporting payment authorization services, the at least one server comprising:
- one or more processors;
- non-transitory computer-readable storage media having computer-executable instructions stored thereon, wherein when executed by the one or more processors the computer-readable instructions cause the one or more processors to—
- receive transaction data corresponding to a plurality of transactions associated with a plurality of transactional entities;
- compute a fraud score for each of the plurality of transactions at least in part by—
- inputting transaction data of the transaction to a classification model and a smart agent associated with the corresponding transactional entity, the smart agent comprising a representation of historical data of the corresponding transactional entity, to generate a first output from the smart agent and a second output from the classification model,
- analyzing the first output and the second output to generate the fraud score for the transaction;
- compute a profitability score for each of the plurality of transactional entities;
- compare the fraud score of each of the plurality of transactions against a first threshold to make a fraud determination for the corresponding transaction;
- compare the profitability score for each of the plurality of transactional entities against a second threshold to make a profitability determination for the corresponding transactional entity;
- and
- for each of the plurality of transactions in which the corresponding fraud determination is positive but the corresponding profitability determination is also positive, issue a transaction approval message.
8 more elements
Non-transitory computer-readable storage media having computer-executable instructions for supporting payment authorization services, wherein when executed by at least one processor the computer-readable instructions cause the at least one processor to:
- receive transaction data corresponding to a plurality of transactions associated with a plurality of transactional entities;
- compute a fraud score for each of the plurality of transactions at least in part by—
- inputting transaction data of the transaction to a classification model and a smart agent associated with the corresponding transactional entity, the smart agent comprising a representation of historical data of the corresponding transactional entity, to generate a first output from the smart agent and a second output from the classification model,
- analyzing the first output and the second output to generate the fraud score for the transaction;
- compute a profitability score for each of the plurality of transactional entities;
- compare the fraud score of each of the plurality of transactions against a first threshold to make a fraud determination for the corresponding transaction;
- compare the profitability score for each of the plurality of transactional entities against a second threshold to make a profitability determination for the corresponding transactional entity;
- and
- for each of the plurality of transactions in which the corresponding fraud determination is positive but the corresponding profitability determination is also positive, issue a transaction approval message.
6 more elements
A computer-implemented method for supporting payment authorization services comprising, via one or more transceivers and/or processors:
- receiving transaction data corresponding to a plurality of transactions associated with a plurality of transactional entities;
- computing a fraud score for each of the plurality of transactions at least in part by—
- inputting transaction data of the transaction to a classification model and a smart agent associated with the corresponding transactional entity, the smart agent comprising a representation of historical data of the corresponding transactional entity, to generate a first output from the smart agent and a second output from the classification model,
- analyzing the first output and the second output using to generate the fraud score for the transaction;
- computing a profitability score for each of the plurality of transactional entities;
- comparing the fraud score of each of the plurality of transactions against a first threshold to make a fraud determination for the corresponding transaction;
- comparing the profitability score for each of the plurality of transactional entities against a second threshold to make a profitability determination for the corresponding transactional entity;
- and
- for each of the plurality of transactions in which the corresponding fraud determination is positive but the corresponding profitability determination is also positive, issuing a transaction approval message.
6 more elements
Dependencies are inferred from the claim text.