Methods and systems for performing secure transactions associated with instructions received in natural language form

Patent metadata

Field	Value
Publication number	US-12450577-B2
Application number	US-202318476914-A
Country	US
Kind code	B2
Filing date	Sep 28, 2023
Priority date	May 30, 2020
Publication date	Oct 21, 2025
Grant date	Oct 21, 2025

Abstract

Embodiments provide methods and systems for facilitating scheduled payment transactions to users via an application provided by the server system, the application available on the user device. The method performed by the server system includes receiving a transaction request message from the user device, the transaction request message includes scheduled transaction instruction provided by the user in natural language format. The method includes authenticating the user based on a plurality of user authentication factors received from the user at pre-defined time intervals. In an embodiment, the server system is configured to train a data model using ML algorithms by learning the plurality of user authentication factors based on which the user is automatically authenticated by the server system. Upon successful authentication, the method includes parsing the transaction request message to determine a scheduled transaction instruction. The method includes facilitating, processing of the scheduled transaction based on the scheduled transaction instruction.

First claim

We claim:

1. A computer-implemented method, comprising:
receiving, by a payment application executing on a user device, a voice or text input from a user;
while the user is providing the voice or text input via the payment application, automatically capturing, by the payment application, one or more real-time user authentication factors, the real-time user authentication factors comprising the following: user voice dynamics, a GPS location of the user device, 3D facial photographs of the user, and user typing dynamics;
invoking, by the payment application, an application programing interface (API) exposed by a trained data model residing on a server system hosting the payment application;
extracting, by the payment application, payment transaction information from the voice or text input provided by the user utilizing the trained data model;
generating, via the payment application, an encryption key based on the real-time user authentication factors automatically captured by the payment application;
generating, by the payment application, an encrypted transaction request message by encrypting, with the encryption key, the captured real-time user authentication factors and the extracted payment transaction information;
transmitting, by the payment application, the encrypted transaction request message to the server system;
receiving, by the server system, the encrypted transaction request message from the payment application;
generating, by the trained data model, an authentication key based on a plurality of authentication templates stored on the server system;
decrypting, by the server system, the encrypted transaction request message using the authentication key;
matching, by the server system, the real-time user authentication factors with the plurality of authentication templates;
in response to a successful match, authenticating, by the server system, the user;
generating, by the server system, a payment processing request; and
transmitting, by the server system, the payment processing request to a payment server.

2. The computer-implemented method as claimed in claim 1 , further comprising:
receiving, by the server system from the user device, a plurality of user authentication factors at pre-defined time intervals;
updating, by the server system, the trained data model, based, at least in part, on the plurality of user authentication factors; and
generating, by the server system, the plurality of authentication templates based on the trained data model.

3. The computer-implemented method as claimed in claim 2 , wherein the plurality of user authentication factors comprises:
recorded voice notes of the user;
a GPS location of the user device;
3D facial photographs of the user; and
user typing dynamics.

4. The computer-implemented method as claimed in claim 2 , further comprising training, by the server system, the trained data model using the plurality of user authentication factors.

5. The computer-implemented method as claimed in claim 1 , wherein the real-time user authentication factors further comprises:
user speech factors; and
a video of the user.

6. The computer-implemented method as claimed in claim 1 , further comprising:
parsing the transaction request message; and
determining, from the payment transaction information, a scheduled transaction instruction,
wherein the scheduled transaction instruction further comprises user account details, a scheduled time, a transaction amount, and at least one of the following: recipient account details and a recipient phone number.

7. The computer-implemented method as claimed in claim 6 , wherein transmitting the payment processing request to the payment server comprises transmitting the payment processing request to the payment server at the scheduled time.

8. The computer-implemented method as claimed in claim 6 , wherein the scheduled transaction instruction is in the form of a text input or a voice input.

9. The computer-implemented method as claimed in claim 1 , wherein the step of extracting the payment transaction information from the voice or text input utilizing the trained data model includes using natural language processing (NLP).

10. A system comprising:
a user device having a payment application executing thereon, the payment application configured to:
receive a voice or text input from a user;
while the user is providing the voice or text input, automatically capture one or more real-time user authentication factors, the real-time user authentication factors comprising the following: user voice dynamics, a GPS location of the user device, 3D facial photographs of the user, and user typing dynamics;
invoke an application programing interface (API) exposed by a trained data model residing on a server system hosting the payment application;
extract payment transaction information from the voice or text input provided by the user utilizing the trained data model;
generate an encryption key based on the real-time user authentication factors automatically captured by the payment application;
generate an encrypted transaction request message, including:
encrypting the captured real-time user authentication factors and the extracted payment transaction information; and

transmit the encrypted transaction request message to the server system; and

a server system hosting the payment application, the server system comprising:
a communication interface configured to:
receive the encrypted transaction request message from the payment application,

a memory storing a plurality of authentication templates and the trained data model thereon, the memory comprising executable instructions; and
a processor communicably coupled to the communication interface and the memory, the processor configured to execute the executable instructions, which cause the processor to:
generate, using the trained data model, an authentication key based on the plurality of authentication templates;
decrypt the encrypted transaction request message using the authentication key;
match the real-time user authentication factors with the plurality of authentication templates;
in response to a successful match, authenticate the user;
generate a payment processing request; and
transmit the payment processing request to a payment server.

11. The system as claimed in claim 10 , wherein the processor is further caused to:
receive, from the user device, a plurality of user authentication factors at pre-defined time intervals;
update the trained data model based, at least in part, on the plurality of user authentication factors; and
generate the plurality of authentication templates based on the trained data model.

12. The system as claimed in claim 11 , wherein the plurality of user authentication factors comprises:
recorded voice notes of the user;
a GPS location of the user device;
3D facial photographs of the user; and
user typing dynamics.

13. The system as claimed in claim 11 , wherein the processor is further caused to train the trained data model using the plurality of user authentication factors.

14. The system as claimed in claim 10 , wherein real-time user authentication factors further comprise:
user speech factors; and
a video of the user.

15. The system as claimed in claim 10 , wherein the processor is further caused to:
parse the transaction request message; and
determine, from the payment transaction information, a scheduled transaction instruction,
wherein the scheduled transaction instruction further comprises user account details, a scheduled time, a transaction amount, and at least one

Assignees

• Mastercard International Inc

Inventors

• Bansal Peeyush
• Kumar Avinash
• Mehra Rishabh

Classifications

• G06Q20/40145

  Biometric identity checks · CPC title

• G06Q2220/00

  Business processing using cryptography · CPC title

• G06F40/205

  Parsing · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q20/386

  using messaging services or messaging apps (using mobile network messaging services for payment, e.g. SMS G06Q20/3255) · CPC title