Granted Patent B2

US 12,204,672 · App. 18/050,631 · Granted Jan 21, 2025

Inventors: Luke Jurat (Fairfax, VA); Vuk Radoicic (Belgrade, RS); Stefan Lazovic (Belgrade, RS); Brandon Castro (Jacksonville, FL); Jeffrey G. Poli (Wauwatosa, WI)

Assignee: Fidelity Information Services, LLC

G06F21/6218G06F16/27G06F21/1011G06F21/1014G06F21/1015G06F21/32G06F21/64H04L63/10G06Q20/065G06Q20/326G06Q20/40H04L9/3231H04L63/083H04L63/0861

Abstract

Claims (81)

1. A method for generating dynamic non-fungible tokens, comprising:

receiving an authentication request for a transaction from at least one device associated with a user;

collecting, via one or more data mining applications or web-crawling applications, identification data associated with the at least one device or the user from one or more databases;

processing, via a machine learning model, the identification data associated with the at least one device or the user to generate the dynamic non-fungible tokens;

storing, via the machine learning model, the dynamic non-fungible tokens in a transaction block of a distributed blockchain; and

transmitting the dynamic non-fungible tokens to a digital wallet application to authenticate the transaction, wherein the dynamic non-fungible tokens is transmitted from a first digital wallet application associated with the user to a second digital wallet application with a service provider,

wherein a supervised learning is utilized to train the machine learning model, the supervised learning including:

inputting, in real-time or near real-time, training data into the machine learning model for generating the dynamic non-fungible tokens for storage into the distributed blockchain, wherein the training data includes an input and a correct output;

monitoring, in real-time or near real-time, accuracy of the machine learning model through a loss function; and

adjusting the machine learning model until an identified error is minimized.

2. The method of claim 1 , wherein generating the dynamic non-fungible tokens, further comprises:

cryptographically hashing, via the machine learning model, the identification data;

concatenating, via the machine learning model, each of the hashed identification data in a pre-defined order;

generating and storing, via the machine learning model, a singular hash representing the concatenated hashed identification data; and

generating the dynamic non-fungible tokens representing the singular hash.

3. The method of claim 1 , wherein storing the dynamic non-fungible tokens includes minting the dynamic non-fungible tokens, further comprises:

validating the dynamic non-fungible tokens;

creating a new transaction block in the distributed blockchain for the dynamic non-fungible tokens, and

recording the dynamic non-fungible tokens into the transaction block in the distributed blockchain.

4. The method of claim 1 , wherein collecting the identification data, further comprises:

generating, in a user interface of the at least one device, a notification requesting access to current location of the user; and

comparing the current location of the user to a store location information to perform a location-based verification to authenticate the transaction.

5. The method of claim 4 , further comprising:

generating, in the user interface of the at least one device, a presentation of one or more instructions to align an identification card, position the user's face in a specific position, or a combination thereof;

capturing, via one or more sensors, a plurality of images or videos of the identification card, the user's face, or a combination thereof; and

analyzing, via one or more algorithms, the plurality of images or videos for accuracy.

6. The method of claim 1 , further comprising:

receiving a blockchain address and a verification that the transaction block is recorded in the distributed blockchain; and

monitoring, in real-time or near real-time, the distributed blockchain and at least one transaction on the distributed blockchain that matches the blockchain address.

7. The method of claim 6 , further comprising:

updating metadata associated with the dynamic non-fungible tokens based, at least in part, on the monitoring;

generating a new dynamic non-fungible token based, at least in part, on the updated metadata; and

concatenating the new dynamic non-fungible token to a preceding dynamic non-fungible token on the transaction block of the distributed blockchain.

8. The method of claim 1 , wherein the distributed blockchain stores the dynamic non-fungible tokens as a sequence of transaction blocks, and wherein each of the transaction block is immutably connected to a preceding transaction block by a cryptographic hash function.

9. The method of claim 1 , wherein the dynamic non-fungible tokens are non-fungible cryptographic assets in a standard token format, and wherein the dynamic non-fungible tokens are recorded in a programmatically defined smart contract written to the distributed blockchain.

10. A non-transitory computer readable medium for generating dynamic non-fungible tokens, the non-transitory computer readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform operations, comprising:

receiving an authentication request for a transaction from at least one device associated with a user;

collecting, via one or more data mining applications or web-crawling applications, identification data associated with the at least one device or the user from one or more databases;

processing, via a machine learning model, the identification data associated with the at least one device or the user to generate the dynamic non-fungible tokens;

storing, via the machine learning model, the dynamic non-fungible tokens in a transaction block of a distributed blockchain; and

transmitting the dynamic non-fungible tokens to a digital wallet application to authenticate the transaction, wherein the non-fungible tokens is transmitted from a first digital wallet application associated with the user to a second digital wallet application with a service provider,

wherein a supervised learning is utilized to train the machine learning model, the supervised learning including:

inputting, in real-time or near real-time, training data into the machine learning model for generating the dynamic non-fungible tokens for storage into the distributed blockchain, wherein the training data includes an input and a correct output;

monitoring, in real-time or near real-time, accuracy of the machine learning model through a loss function; and

adjusting the machine learning model until an identified error is minimized.

11. The non-transitory computer readable medium of claim 10 , wherein generating the dynamic non-fungible tokens, further comprises:

cryptographically hashing, via the machine learning model, the identification data;

concatenating, via the machine learning model, each of the hashed identification data in a pre-defined order;

generating and storing, via the machine learning model, a singular hash representing the concatenated hashed identification data; and

generating the dynamic non-fungible tokens representing the singular hash.

12. The non-transitory computer readable medium of claim 10 , wherein collecting the identification data, further comprises:

generating, in a user interface of the at least one device, a notification requesting access to current location of the user; and

comparing the current location of the user to a store location information to perform a location-based verification to authenticate the transaction.

13. The non-transitory computer readable medium of claim 12 , further comprising:

generating, in a user interface of the at least one device, a presentation of one or more instructions to align an identification card, position the user's face in a specific position, or a combination thereof;

capturing, via one or more sensors, a plurality of images or videos of the identification card, the user's face, or a combination thereof; and

analyzing, via one or more algorithms, the plurality of images or videos for accuracy.

14. The non-transitory computer readable medium of claim 10 , further comprising:

receiving a blockchain address and a verification that the transaction block is recorded in the distributed blockchain; and

monitoring, in real-time or near real-time, the distributed blockchain and at least one transaction on the distributed blockchain that matches the blockchain address.

15. The non-transitory computer readable medium of claim 14 , further comprising:

updating metadata associated with the dynamic non-fungible tokens based, at least in part, on the monitoring;

generating a new dynamic non-fungible token based, at least in part, on the updated metadata; and

concatenating the new dynamic non-fungible token to a preceding dynamic non-fungible token on the transaction block of the distributed blockchain.

16. A system for generating dynamic non-fungible tokens, comprising:

one or more processors; and

at least one non-transitory computer readable medium storing instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

receiving an authentication request for a transaction from at least one device associated with a user;

collecting, via one or more data mining applications or web-crawling applications, identification data associated with the at least one device or the user from one or more databases;

processing, via a machine learning model, the identification data associated with the at least one device or the user to generate the dynamic non-fungible tokens;

storing, via the machine learning model, the dynamic non-fungible tokens in a transaction block of a distributed blockchain; and

transmitting the dynamic non-fungible tokens to a digital wallet application to authenticate the transaction, wherein the non-fungible tokens is transmitted from a first digital wallet application associated with the user to a second digital wallet application with a service provider,

wherein a supervised learning is utilized to train the machine learning model, the supervised learning including:

inputting, in real-time or near real-time, training data into the machine learning model for generating the dynamic non-fungible tokens for storage into the distributed blockchain, wherein the training data includes an input and a correct output;

monitoring, in real-time or near real-time, accuracy of the machine learning model through a loss function; and

adjusting the machine learning model until an identified error is minimized.

17. The system of claim 16 , wherein generating the dynamic non-fungible tokens, further comprises:

cryptographically hashing, via the machine learning model, the identification data;

concatenating, via the machine learning model, each of the hashed identification data in a pre-defined order;

generating and storing, via the machine learning model, a singular hash representing the concatenated hashed identification data; and

generating the dynamic non-fungible tokens representing the singular hash.

Assignments (1)

Continuity (2)

References Cited (30)