Automated multi-party event and transaction decisioning system

Patent metadata

Field	Value
Publication number	US-11868973-B2
Application number	US-202117406688-A
Country	US
Kind code	B2
Filing date	Aug 19, 2021
Priority date	Aug 19, 2021
Publication date	Jan 9, 2024
Grant date	Jan 9, 2024

Abstract

Methods, systems, devices, and computer-readable media for detecting multi-party events and transactions are provided. User data may be monitored to detect data associated with an event involving multiple individuals, such as by identifying transaction data associated with certain types of merchants and/or scheduling, calendar, or correspondence data indicative of an event. The data may be further analyzed to identify a date, location, and/or parties associated with the event. A multi-party event may be generated. The user data may continue to be monitored to identify transactions associated with multiple parties and occurring during a time and/or at a location of the event. At a conclusion of the event, the transactions may be aggregated and an optimal payment scheme may be determined for settlement of the transactions between the parties. In accordance with the determined payment scheme, delegation of portions of the aggregated transactions may be initiated for settlement amongst the parties.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving, by a computing device, first data comprising a plurality of first transactions associated with a first user;
detecting, in the first data, a first transaction, of the plurality of first transactions, associated with one or more merchant types associated with multi-party events;
retrieving, in response to detecting the first transaction and from a first device associated with the first user, second data comprising correspondence data or calendar data associated with the first user;
analyzing the second data to identify at least one correspondence or at least one calendar entry associated with the first transaction;
identifying, based on scanning the at least one correspondence or the at least one calendar entry, a location and a time associated with the first transaction, and a second user associated with the first transaction;
generating, by the computing device, based on identifying the location and the time associated with the first transaction and using a first machine learning model, a multi-party event comprising information identifying the identified location and the identified time, wherein the first machine learning model is trained based on training data comprising:
pre-labeled transactions indicating whether a training transaction record is a multi-party transaction or an individual transaction,
a transaction amount of the training transaction record,
a date and time associated with the training transaction record,
a type of merchant associated with the training transaction record,
a class of merchant associated with the training transaction record, and
a location of merchant associated with the training transaction record;

generating, by the computing device, a payment token for the multi-party event, wherein the payment token comprises a URL link to a script configured to access information on one or more delegation requests associated with the multi-party event;
storing, in a database, the multi-party event and the payment token;
associating, in the database, the first transaction with the multi-party event;
associating, in the database, the first user as a first party to the multi-party event and the second user as a second party to the multi-party event;
transmitting, to the first device associated with the first user, a notification indicating identification of the multi-party event; and
transmitting, to a second device associated with the second user, a delegation message comprising the payment token in an encrypted format.

2. The computer-implemented method of claim 1 , further comprising:
receiving, from the first device associated with the first user, an indication that the first user is located proximate to the identified location during a time frame encompassing the identified time,
wherein generating the multi-party event is further based on receiving the indication that the first user is located proximate to the identified location during the time frame encompassing the identified time.

3. The computer-implemented method of claim 1 , wherein the one or more merchant types comprise a transportation provider,
wherein the first transaction comprises a transaction for a transportation-related purchase, and
wherein the at least one correspondence comprises an email comprising a receipt for the transportation-related purchase.

4. The computer-implemented method of claim 1 , further comprising:
receiving information identifying the second user associated with the first transaction;
obtaining, from the first device associated with the first user, contact information for the second user; and
transmitting, to the second user and using the contact information, a request for the second user to share transactions of the second user with the multi-party event.

5. The computer-implemented method of claim 1 , further comprising:
associating, in the database, the second user as the second party to the multi-party event.

6. The computer-implemented method of claim 5 , further comprising:
receiving, by the computing device, additional data comprising a plurality of second transactions associated with the first user and the second user;
detecting, in the additional data, one or more transactions having a location and a time that correspond to the identified location and the identified time associated with the multi-party event; and
associating, in the database, the one or more transactions with the multi-party event.

7. The computer-implemented method of claim 6 , further comprising:
retrieving, in response to detecting the one or more transactions and from one or more of the first device associated with the first user or the second device associated with the second user, user data associated with the first user or the second user;
analyzing the user data to identify one or more third users associated with the one or more transactions; and
associating, in the database, the one or more third users as third parties to the multi-party event.

8. The computer-implemented method of claim 1 , further comprising:
determining, based on the identified time associated with the multi-party event, that the multi-party event has concluded;
retrieving, from the database, transactions and one or more parties associated with the multi-party event;
generating, based on an aggregation of transaction amounts of the transactions associated with the multi-party event, an optimized payment scheme for payment to the one or more parties associated with the multi-party event;
generating, based on the optimized payment scheme, the one or more delegation requests, wherein each delegation request comprises information indicating:
a party from the multi-party event identified as a payor party,
a party from the multi-party event identified as a payee party, and
an amount to be paid by the payor party to the payee party; and

for each of the one or more delegation requests, transmitting the delegation request to a corresponding payor party.

9. 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:
receive first data comprising a plurality of first transactions associated with a first user;
detect, in the first data, a first transaction, of the plurality of first transactions, associated with one or more merchant types associated with multi-party events;
retrieve, in response to detecting the first transaction and from a first device associated with the first user, second data comprising correspondence data or calendar data associated with the first user;
analyze the second data to identify at least one correspondence or at least one calendar entry associated with the first transaction;
identify, based on scanning the at least one correspondence or the at least one calendar entry, a location and a time associated with the first transaction, and a second user associated with the first transaction;

generate, based on identifying the location and the time associated with the first transaction and using a first machine learning model, a multi-party event comprising information identifying the identified location and the identified time wherein the first machine learning model is trained based on training data comprising:
pre-labeled transactions indicating whether a training transaction record is a multi-party transaction or an individual transaction,
a transaction amount of the training transaction record,
a date and time associated with the training transaction record,
a type of merchant associated with the training transaction record,
a class of merchant associated with the training transaction record,

Assignees

• Capital One Services Llc

Inventors

• Bantanur Vijay Sahebgouda
• Balasubramanian Muralidharan
• Dallen Julie

Classifications

• G06Q20/047 Primary

  using payment protocols involving electronic receipts · CPC title

• G06Q20/123

  Shopping for digital content · CPC title

• G06Q20/347

  Passive cards · CPC title

• G06Q20/40155

  for triggering transactions · CPC title