# US 12,762,591 B2 · System and method for real-time generation and optimization of personalized cause-aligned content

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,762,591 B2
- Document ID: US12762591B2
- Status: Granted
- Kind: B2
- Application number: 19/285,297
- Filed: 2025-07-30
- Published: 2026-02-12
- Granted: 2026-10-13
- Earliest claimed: 2024-08-08
- Assignees: Love Little Light LLC
- Applicants: Love Little Light LLC
- Inventors: James Scott; Ami Shah
- Web page: https://patentgrep.com/patent/US12762591B2

## CPC

- G06Q 30/0276: {Advertisement creation}
- G06N 20/00: Machine learning
- G06Q 30/0246: {Traffic}
- G06Q 30/0271: {Personalized advertisement}
- G06Q 30/0279: Fundraising management
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present disclosure provides a computer-implemented system and method for creating and serving cause-aligned digital personalized content that convert user engagement into real-time brand-funded donations. The system employs machine learning algorithms to match brands with compatible causes and relevant users, generating personalized advertisements in real-time. A key feature is the interactive AdsUp button, which triggers micro-donations from brands to causes based on user engagement. The system incorporates a continuous learning module that refines the machine learning model using real-time user engagement data, ensuring ongoing improvement in targeting personalized content and effectiveness.

## Claims

### Claim 1

A method for real-time generation of personalized content for a user, the method comprising:
receiving a request comprising a brand identifier and a user identifier;
retrieving, from a pre-computed low-latency lookup table, a plurality of cause records associated with the brand identifier, wherein each cause record in the low-latency lookup table is associated with a brand-cause compatibility score;
computing, using a multilayered machine learning model, a plurality of cause-user relevance scores, wherein each cause-user relevance score represents a relevance between the user identifier and a respective cause record from the retrieved plurality of cause records, wherein the plurality of cause-user relevance scores is computed by:
receiving, via a sensory layer of the multilayered machine learning model, data associated with one or more brands and one or more cause, wherein the received data includes textual data and image data;
processing the textual data and image data to determine context by:
tokenizing the textual data into one or more tokens, generating first feature vectors for the one or more tokens by iteratively refining token embeddings through multiple self-attention layers and feed-forward neural network layers of a transformer-based natural language model;
dividing the image data into fixed-size patches, and generating second feature vectors for the fixed-size patches by encoding patch embeddings through multiple self-attention layers of a Vision Transformer (ViT); and
combining the first feature vectors and the second feature vectors into a multimodal context representation;
executing, via a cognitive layer of the multilayered machine learning model, an Audience Response Model comprising a long short-term memory (LSTM) network with an attention mechanism on the multimodal context representation, historical audience response sequences associated with the user identifier, and prior engagement metrics corresponding to similar brand-cause campaigns to generate an engagement likelihood representation indicating predicted audience interaction with each respective cause record;
generating, by an alignment scoring model via an executive layer of the multilayered machine learning model, a brand-cause compatibility score based on the engagement likelihood representation;
selecting a target cause record from the plurality of cause records, wherein the target cause record has a highest cause-user relevance score among the plurality of cause-user relevance scores, and wherein the highest cause-user relevance score exceeds a first predetermined threshold;
generating, in real-time, the personalized digital content for the user identifier, by combining brand related content associated with the brand identifier and cause related content associated with the target cause record;
transmitting the personalized digital content for rendering on a device associated with the user identifier;
tracking user engagement data associated with a personalized advertisement; and
continuously training the multilayered machine learning model based on the user engagement data to refine at least the cause-user relevance scores and the brand-cause compatibility score.

### Claim 2 (depends on 1)

The method of claim 1, wherein the pre-computed low-latency lookup table is generated by analysing historical brand-cause alignment data.

### Claim 3 (depends on 1)

The method of claim 1, wherein the personalized advertisement includes an interactive element that, when activated, triggers a donation to a charitable organization associated with the target cause record.

### Claim 4 (depends on 3)

The method of claim 3, further comprising processing, in real-time, a transaction to execute the donation triggered by activation of the interactive element.

### Claim 5 (depends on 1)

The method of claim 1, wherein the user engagement data includes at least one of: click-through rates, time spent viewing the advertisement, social media shares, and donation amounts.

### Claim 6 (depends on 1)

The method of claim 1, further comprising dynamically adjusting the first predetermined threshold and a second predetermined threshold based on aggregate user engagement data across multiple users and brands.

### Claim 7 (depends on 1)

The method of claim 1, wherein the multi-layered machine learning model configured to compute the brand-cause compatibility score, the model comprising:
the sensory layer comprising:
the transformer-based natural language model configured to generate brand embeddings and cause embeddings from textual inputs associated with a brand and a cause; and
the vision transformer configured to extract features from visual content associated with the brand and the cause;
the cognitive layer comprising:
a cause analysis model implemented using a graph neural network and configured to process cause-related data and cause inter-relationships; and
the audience response model comprising a long short-term memory (LSTM) network with attention and configured to predict engagement likelihood based on historical audience behaviour; and
the executive layer comprising:
the alignment scoring model configured to receive the brand embeddings, cause embeddings, and outputs from the cognitive layer, and to compute the brand-cause compatibility score; and
an impact prediction model configured to estimate user engagement based on prior campaign performance and contextual variables,
wherein the multi-layered machine learning model is trained using labeled brand-cause data and real-time engagement metrics, and
wherein the trained multi-layered machine learning model is deployable within a content personalization engine to identify and rank causes aligned with the brand for digital content generation.

### Claim 8

A system for real-time generation of personalized content for a user, the system comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the system to:
receive a request comprising a brand identifier and a user identifier;
retrieve, from a pre-computed low-latency lookup table, a plurality of cause records associated with the brand identifier, wherein each cause record in the low-latency lookup table is associated with a brand-cause compatibility score;
compute, using a multilayered machine learning model, a plurality of cause-user relevance scores, wherein each cause-user relevance score represents a relevance between the user identifier and a respective cause record from the retrieved plurality of cause records, wherein the plurality of cause-user relevance scores is computed by:
receiving, via a sensory layer of the multilayered machine learning model, data associated with one or more brands and one or more causes, wherein the received data includes textual data and image data;
processing the textual data and image data to determine context by:
tokenizing the textual data into one or more tokens, generating first feature vectors for the one or more tokens by iteratively refining token embeddings through multiple self-attention layers and feed-forward neural network layers of a transformer-based natural language model;
dividing the image data into fixed-size patches, generating second feature vectors for the fixed-size patches by encoding patch embeddings through multiple self-attention layers of a Vision Transformer (ViT); and
combining the first feature vectors and the second feature vectors into a multimodal context representation;
executing, via a cognitive layer of the multilayered machine learning model, an Audience Response Model comprising a long short-term memory (LSTM) network with an attention mechanism on the multimodal context representation, historical audience response sequences associated with the user identifier, and prior engagement metrics corresponding to similar brand-cause campaigns to generate an engagement likelihood representation indicating predicted audience interaction with each respective cause record; and
generating, by an alignment scoring model via an executive layer of the multilayered machine learning model, a brand-cause compatibility score based on the engagement likelihood representation;
select a target cause record from the plurality of cause records, wherein the target cause record has a highest cause-user relevance score among the plurality of cause-user relevance scores, and wherein the highest cause-user relevance score exceeds a predetermined threshold;
generate, in real-time, the personalized digital content for the user identifier, by combining brand-related content associated with the brand identifier and cause-related content associated with the target cause record;
transmit a personalized advertisement for rendering on a device associated with the user identifier;
track user engagement data associated with the personalized digital content; and
continuously train the multilayered machine learning model based on the user engagement data to refine at least the cause-user relevance scores and the brand-cause compatibility score.

### Claim 9 (depends on 8)

The system of claim 8, wherein the pre-computed low-latency lookup table is generated by analysing historical brand-cause alignment data.

### Claim 10 (depends on 8)

The system of claim 8, wherein the personalized advertisement includes an interactive element that, when activated, triggers a donation to a charitable organization associated with the target cause record.

### Claim 11 (depends on 8)

The system of claim 8, wherein the instructions further cause the processor to process, in real-time, a transaction to execute the donation triggered by activation of the interactive element.

### Claim 12 (depends on 8)

The system of claim 8, wherein the user engagement data includes at least one of: click-through rates, time spent viewing the advertisement, social media shares, and donation amounts.

### Claim 13 (depends on 8)

The system of claim 8, wherein the instructions further cause the processor to dynamically adjust the predetermined threshold based on aggregate user engagement data across multiple users and brands.