Transaction exchange platform with a validation microservice for validating transactions before being processed

Patent metadata

Field	Value
Publication number	US-2023153825-A1
Application number	US-202318149493-A
Country	US
Kind code	A1
Filing date	Jan 3, 2023
Priority date	Dec 20, 2019
Publication date	May 18, 2023
Grant date	—

Abstract

Aspects described herein may relate to a transaction exchange platform using a streaming data platform (SDP) and microservices to process transactions according to review and approval workflows. The transaction exchange platform may receive transactions from origination sources, which may be added to the SDP as transaction objects. As the transactions are received, the transactions may be analyzed to detect duplicate transactions and/or errors in the transactions. The transaction exchange platform may take steps to remediate transactions that are recognized as duplicates or predicted to generate one or more errors. Similarly, the transaction exchange platform may take steps to remediate transactions that are rejected by a clearinghouse.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
training a predictive model to identify payment transactions that have a likelihood of failing due to one or more errors, wherein a dataset for training the predictive model comprises:
transaction details associated with a plurality of payment transactions that failed; and
a reason each of the plurality of transactions failed;
receiving, by a screening microservice, a first transaction object corresponding to a first payment transaction, wherein the first transaction object comprises first transaction details; determining, by the screening microservice and using the predictive model, whether the first transaction details indicate that processing of the first transaction object is likely to fail, wherein determining whether the transaction details indicate that processing of the first transaction object is likely to fail is based on a first likelihood of failing exceeding a threshold; based on a determination that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail, determining a corrective action to reduce the first likelihood that processing of the first transaction object is going to fail; sending, by the screening microservice to a first user device, an indication that processing of the first transaction object is likely to fail and the corrective action; receiving, by the screening microservice from the first user device, a response indicating acceptance of the corrective action; updating, by the screening microservice and based on the response indicating acceptance of the corrective action, addenda data associated with the first transaction object; and adding, by the screening microservice, the first transaction object with updated addenda data to a streaming data platform, wherein adding the first transaction object to the streaming data platform comprises setting a current workflow stage of the first transaction object to an initialization stage.

2 . The computer-implemented method of claim 1 , wherein the determination that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail further comprises:
determining, by the screening microservice, that a first account associated with a payor does not comprise sufficient funds to cover a first transaction associated with the first transaction object, wherein the corrective action comprises transferring funds from a second account to a first account.

3 . The computer-implemented method of claim 1 , wherein the determination that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail further comprises at least one of:
determining, by the screening microservice, that the first transaction details do not identify an account associated with a payor, wherein the corrective action comprises updating information associated with the account to correctly identify an account associated with the payor; determining, by the screening microservice, that the first transaction details do not identify an account associated with a payee, wherein the corrective action comprises updating information associated with the account to correctly identify an account associated with the payee; or determining, by the screening microservice, that the first transaction details do not comport with a first workflow for processing the first transaction object, wherein the corrective action comprises changing a workflow type of the first transaction object to a second workflow.

4 . The computer-implemented method of claim 1 , wherein the transaction details associated with the plurality of payment transactions that failed comprise at least one of: an identification of a payor, an identification of a payee, a dollar amount of a transaction, a time of the transaction, an identification of a type of platform on which the transaction occurs, a type of product transaction code, or core attributes of the transaction.

5 . The computer-implemented method of claim 1 , wherein the reason the transaction failed comprises a return code.

6 . The computer-implemented method of claim 1 , wherein the predictive model comprises at least one of: k-means algorithm, affinity propagation algorithm, mean-shift algorithm, spectral clustering algorithm, Ward hierarchical clustering algorithm, agglomerative clustering algorithm, density-based spatial clustering of applications with noise (DBSCAN) algorithm, Gaussian mixtures algorithm, Birch algorithm, or shared nearest neighbors algorithm.

7 . A computing device comprising:
at least one processor; and memory storing instructions that, when executed by the at least one processor, cause the computing device to:
train a predictive model to identify payment transactions that have a likelihood of failing due to one or more errors, wherein a dataset for training the predictive model comprises:
transaction details associated with a plurality of payment transactions that failed; and
a reason each of the plurality of transactions failed;

receive, by a screening microservice, a first transaction object corresponding to a first payment transaction, wherein the first transaction object comprises first transaction details;
determine, by the screening microservice and using the predictive model, whether the first transaction details indicate that processing of the first transaction object is likely to fail, wherein determining whether the transaction details indicate that processing of the first transaction object is likely to fail is based on a first likelihood of failing exceeding a threshold;
based on a determination that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail, determine a corrective action to reduce the first likelihood that processing of the first transaction object is going to fail;
send, by the screening microservice to a first user device, an indication that processing of the first transaction object is likely to fail and the corrective action;
receive, by the screening microservice from the first user device, a response indicating acceptance of the corrective action;
update, by the screening microservice and based on the response indicating acceptance of the corrective action, addenda data associated with the first transaction object; and
add, by the screening microservice, the first transaction object with updated addenda data to a streaming data platform, wherein adding the first transaction object to the streaming data platform comprises setting a current workflow stage of the first transaction object to an initialization stage.

8 . The computing device of claim 7 , wherein instructions for determining that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail cause the computing device to:
determine, by the screening microservice, that a first account associated with a payor does not comprise sufficient funds to cover a first transaction associated with the first transaction object, wherein the corrective action comprises transferring funds from a second account to a first account.

9 . The computing device of claim 7 , wherein instructions for determining that the first transaction details indicate the first likelihood that processing of the first transaction object is going to fail cause the computing device to:
determine, by the screening microservice, that the first transaction details do not identify an account associated with a payor, wherein the corrective action comprises updating information associated with the accou

Assignees

• Capital One Services Llc

Inventors

• Srivastava Nishant
• Barnum Eric K
• Ramaraj Suresh Chander
• Smith Eric
• Lee Earle Michael
• Bhawprita Kumari

Classifications

• G06N20/00

  Machine learning · CPC title

• G06F11/0751

  Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06) · CPC title

• G06F2201/81

  Threshold · CPC title

• G06F11/1474

  in transactions (updating of structured data in databases G06F16/23) · CPC title

• G06F2201/87

  Monitoring of transactions · CPC title