Granted Patent B1

US 12,614,213 · App. 18/386,369 · Granted Apr 28, 2026

Inventors: Rachel Alexandria Miller (Fort Worth, TX); Holly Renee Homer (Flower Mound, TX)

G06Q30/0276G06F16/972G06F40/186

Abstract

Claims (70)

1 . A digital content generation platform comprising:

a control server comprising one or more processors, an interface module controlling a preview interface for display on user devices, a product generation module, a feedback and training module, a content data module, and a computer storage device upon which is stored at least a structure library, user preference data from a plurality of users, a content database, and non-transitory computer-readable media comprising computer-executable instructions that, when executed by the one or more processors, cause the digital content generation platform to:

communicate, by the control server, with the user devices of the plurality of users;

receive and associate, by the preview interface, a desired digital product type, a classification input identifying one or more classifiers from the content database, and a structure input identifying at least one structure set comprising structure elements from the structure library, wherein the structure input specifies an initial product layout of the desired digital product type, and each structure element comprises one or more content fields defined by the one or more classifiers within the initial product layout;

match, by the content data module, the one or more content fields identified by the structure input to a plurality of content elements in the content database, and create a content variation pool from the matched one or more content fields and the plurality of content elements;

assign, by the content data module, a favorability parameter to the content variation pool by aggregating and analyzing the user preference data comprising user feedback about individual content fields used for populating individual content elements in previous product layouts, and store the favorability parameter in the content database;

render, by the product generation module, a draft product layout, wherein to render the draft product layout, the execution of the computer-executable instructions by the one or more processors further causes the product generation module to:

retrieve the matched one or more content fields and the plurality of content elements from the content variation pool in the content database, and populate the draft product layout with the matched one or more content fields and the plurality of content elements within the structure elements defined by the structure input;

cause the interface module to display on the preview interface a content preview for each content field within the draft product layout, wherein the product generation module renders an associated content field populated with an associated content element of the content variation pool; and

upon receipt, by the product generation module, of a refresh instruction, cause the interface module to display on the preview interface a refreshed content preview, wherein the product generation module re-renders the draft product layout by populating the content preview with a different content element selected from the content variation pool, and wherein the product generation module selects the different content element from the content variation pool by comparing the favorability parameters of each of the content elements within the content variation pool;

update, by the feedback and training module, the favorability parameter of each of the content elements by increasing the favorability parameter when a corresponding content preview is approved, or by decreasing the favorability parameter when the corresponding content preview is refreshed; and

render, by the product generation module, a final product layout upon receipt, from the preview interface, of a user approval of one or more of the content previews by populating each of the content fields with a content element associated with the corresponding approved content preview, thereby creating a completed digital product.

2 . The digital content generation platform of claim 1 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

present, by the preview interface to the user, a plurality of structure previews, each of the structure previews showing the draft product layout incorporating the structure elements of one of the structure sets associated with the one or more classifiers defined by the classification input;

receive, by the preview interface, a user selection of one of the structure previews, causing the structure input to define the structure set associated with the selected structure preview,

wherein each of the structure sets has a favorability parameter relative to the classifiers with which the structure set is associated;

adjust, by the feedback and training module, the favorability parameter of each of the structure sets by increasing the favorability parameter when the corresponding structure preview is selected; and

select, by the product generation module, the structure set to populate each structure preview by comparing the favorability parameters of each of the structure sets associated with the one or more classifiers defined by the classification input.

3 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

display, by the preview interface, a multi-combination preview associated with one of the content fields within the draft product layout, the multi-combination preview providing a simultaneous visual comparison of at least two of the content elements within the content variation field of the content field, each of the content previews within the multi-combination preview showing the content field populated with a different one of the content elements of the content variation pool.

4 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

order, by the product generation module, the content fields associated with one of the structure elements within the draft product layout in a sequence and link the content fields to a content combination chain,

wherein each of the content elements associated with the content combination chain has a content combination chain favorability parameter, the content combination chain favorability parameter indicating how favorable the content element is relative to the content elements of the content variation fields of the other content fields which precede or follow the content element within the content combination chain; and

populate, by the product generation module, the content previews for each of the content fields associated with the content combination chain by selecting the content elements to maximize the content combination chain favorability parameters thereof.

5 . The digital content generation platform of claim 4 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

regenerate, by the product generation module, at least one of the content previews of one of the content fields associated with the content combination chain when the content preview of the content field preceding the content preview in the sequence is regenerated.

6 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

prior to the user entering the structure input, identify, by the product generation module, potential structure sets associated with the one or more classifiers defined by the classification input;

identify, by the product generation module, the content fields associated with each of the structure elements of the potential structure sets;

retrieve, by the product generation module, the content elements which form the content variation pool of each of the identified content fields;

store, by the product generation module to the computer storage device, the content elements as pregenerated combinations for each of the identified content fields, wherein the pregenerated combinations are stored within a temporary storage of the computer storage device; and

retrieve, by the preview interface, a corresponding pregenerated combination from the temporary storage when populating, by the product generation module, the content previews of the content fields within the draft product layout.

7 . The digital content generation platform of claim 2 , wherein:

the control server further comprises a machine learning module having a generative artificial intelligence (GAI) application, wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

receive, by the preview interface, a new classifier in a new classification input from the user, wherein the new classifier is not found within the content database;

determine, by the product generation module, a plurality of candidate structure sets and associate the plurality of candidate structure sets with the new classifier;

populate, by the preview interface, new structure previews with the plurality of candidate structure sets;

receive, by the preview interface, a new structure input by user selection of one of the new structure previews embodying one of the candidate structure sets;

render, by the product generation module, the draft product layout with the content fields of the one of the candidate structure sets defined by the new structure input;

send, by the product generation module, a content element request to the machine learning module to generate a plurality of unvalidated content elements to form the content variation pool for each of the content fields;

in response to the content element request, generate a prompt, by the machine learning module, wherein the prompt is formatted to instruct the GAI application to generate each of the unvalidated content elements, the prompt containing a classifier instruction identifying the new classifier;

select, by the product generation module, one of the unvalidated content elements to populate a new content preview for each of the content fields from within the content variation pool;

receive, by the preview interface, a new user approval of one or more of the new content previews showing the one of the unvalidated content elements;

in response to receiving the new user approval, populate, by the product generation module, the content fields of the draft product layout using the one of the unvalidated content elements associated with the corresponding approved content preview;

receive, by the preview interface, an indication to refresh the one of the unvalidated content elements by user selection of the corresponding approved content preview; and

in response to receiving the indication to refresh the one of the unvalidated content elements, populate, by the product generation module, another new content preview with a different unvalidated content element from the content variation pool.

8 . The digital content generation platform of claim 7 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

store, by the content data module, the unvalidated content elements within an unvalidated content store, store the new classifier as an unvalidated classifier within the unvalidated content store, and associate the unvalidated content elements with the unvalidated classifier and the structure elements of a corresponding candidate structure set;

retrieve, by the product generation module, the unvalidated content elements from the unvalidated content store when a subsequent classification input by one of the users selects the unvalidated classifier,

wherein each of the unvalidated content elements has a favorability parameter; and

increase, by the feedback and training module, the favorability parameter of a corresponding unvalidated content element when the corresponding content preview is approved, or decrease the favorability parameter of the corresponding unvalidated content element when the corresponding content preview is refreshed.

9 . The digital content generation platform of claim 8 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

transfer, by the content data module, the corresponding unvalidated content element from the unvalidated content store to the content database when the favorability parameter of the corresponding unvalidated content element exceeds a positive threshold.

10 . The digital content generation platform of claim 8 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

associate, by the product generation module, one or more of the content fields within one of the candidate structure sets defined by the structure input with a content instruction to instruct the machine learning module to generate a prompt; and

combine, by the machine learning module, the content instruction and the classifier instruction to generate the prompt to instruct the GAI application to generate each of the unvalidated content elements associated with the one or more content fields.

11 . The digital content generation platform of claim 8 , wherein:

the classification input defines at least one of the classifiers within the content database in addition to the new classifier to form a classifier grouping; and

the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

retrieve, by the machine learning module, an example instruction corresponding to a content element associated with the one or more classifiers within the content database within the classifier grouping; and

generate, by the machine learning module, the prompt for one of the unvalidated content elements by combining the classifier instruction with the example instruction, causing the GAI application to pattern generated unvalidated content after the example instruction.

12 . The digital content generation platform of claim 8 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

pre-generate, by the product generation module, the unvalidated content elements by sending the content element request to generate the unvalidated content elements to form the content variation pool associated with each of the candidate structure sets once the new classifier is defined and prior to the user entering the structure input.

13 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

batch, by the product generation module, a plurality of templates into the draft product layout, wherein the draft product layout comprises a plurality of batches, each batch has a distinct set of classifiers or sub-classifiers, and the structure elements of a digital product are contained within one of the batches, and wherein the batching provides an organizational structure to the digital product with distinct components or sections; and

receive, by the interface module, a unique classification input and a unique structure input from the user for each of the batches.

14 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

present, by the preview interface, each structure preview or each content preview in an expanded state.

15 . The digital content generation platform of claim 2 , wherein the computer-executable instructions, when executed by the one or more processors, further cause the digital content generation platform to:

output, by the product generation module, the completed digital product as a downloadable file.

Assignments (2)

References Cited (32)