# US 2021/0344621 A1 · Systems and methods for updating creatives generation models

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

- Number: US 2021/0344621 A1
- Document ID: US20210344621A1
- Status: Published application
- Kind: A1
- Application number: 17/200,572
- Filed: 2021-03-12
- Published: 2021-11-04
- Earliest claimed: 2020-03-04
- Assignees: Capital One Services, LLC
- Applicants: Capital One Services, LLC
- Inventors: Kirankumar KULKARNI; Savio Joseph DARIVEMULA; Anil KONDURU; Gunjan PATEL; Jie SHEN; Kelly L. BIRCH; Patrick James MANION
- Web page: https://patentgrep.com/patent/US20210344621A1

## CPC

- H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06N 5/04: Inference or reasoning models
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Disclosed embodiments provide systems and methods related to updating creatives generation models. The system may include at least one memory unit for storing instructions and at least one processor configured to execute the instructions to perform operations. The operations may include receiving a feedback indication concerning an email message sent to a target, the email message constructed using a first template and associated with a first context, the feedback indication including a recommendation identifier; updating, in response to receiving the feedback indication, a feedback value for the email message stored in a delay buffer; obtaining the updated feedback value upon satisfaction of a time delay condition; updating, using the updated feedback data and the recommendation identifier, a machine learning model configured to recommend templates based on contexts; and constructing and providing a second email message using a second template recommended by the updated machine learning model for a second context.

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

A system comprising:
at least one memory containing instructions; and
at least one processor configured to execute the instructions to perform operations comprising:
receiving a request to construct a first email message, the request including an identifier of a first target;
receiving a feedback indication concerning the first email message sent to the first target, the first email message constructed using a first template and associated with a first context;
updating, in response to receiving the feedback indication, a feedback value for the first email message;
updating, using the updated feedback value and the identifier, a machine learning model configured to recommend templates based on contexts; and
constructing and providing a second email message using a second template recommended by the updated machine learning model for a second context.

### Claim 22 (depends on 21)

The system of claim 21, wherein the feedback indication indicates selection of a link in the first email message.

### Claim 23 (depends on 21)

The system of claim 21, wherein the first context includes descriptive data for a first target and interaction data for the first target.

### Claim 24 (depends on 21)

The system of claim 21, wherein updating a feedback value further comprises changing a negative feedback value into a positive feedback value.

### Claim 25 (depends on 21)

The system of claim 21, wherein the operations further comprise obtaining the updated feedback value upon satisfaction of time delay.

### Claim 26 (depends on 21)

The system of claim 21, wherein the time delay condition is satisfied when a time between 8 and 30 hours has elapsed since mailing of the first email message.

### Claim 27 (depends on 21)

The system of claim 21, wherein the machine learning model is updated to increase a probability of recommending the first template based on the first context.

### Claim 28 (depends on 21)

The system of claim 21, wherein updating a feedback value further comprises storing the first email message in a delay buffer.

### Claim 29 (depends on 21)

The system of claim 21, wherein constructing and providing the second email message further comprises:
receiving a request to generate the second email message, the request including an identifier of the machine learning model and an identifier of a second target;
retrieving, from at least one database, information for the second target, using the identifier of the second target;
applying at least a portion of the information to the machine learning model to generate a template recommendation;
constructing the second email message using the template recommendation; and
providing the constructed second email message.

### Claim 30 (depends on 29)

The system of claim 29, wherein constructing the second email message using the template recommendation further comprises retrieving, from a database storing a set of templates, the second template using the template recommendation.

### Claim 31

The system of claim of 29, wherein the operations further comprise retrieving a campaign definition that specifies the at least a portion of the information applied to the machine learning model.

### Claim 32 (depends on 29)

The system of claim 29, wherein the operations further comprise storing a negative feedback value for the second email message in a delay buffer with a delay time.

### Claim 33 (depends on 32)

The system of claim 32, wherein the negative feedback value is stored in the delay buffer in response to receiving a delivery indication for the first email.

### Claim 34

A system comprising:
at least one memory containing instructions; and
at least one processor configured to execute the instructions to perform operations comprising:
receiving a request to generate a first email message, the request including an identifier of a first target;
retrieving, from at least one database, information for the first target, using the identifier of the first target;
applying at least a portion of the information to a machine learning model to generate a template recommendation;
constructing the first email message using the template recommendation; and
providing the constructed first email message.

### Claim 35 (depends on 34)

The system of claim 34, wherein the operations further comprise storing a negative feedback value for the first email message in a delay buffer with a delay time.

### Claim 36 (depends on 34)

The system of claim 34, wherein the operations further comprise retrieving a campaign definition that specifies the at least a portion of the information applied to the machine learning model.

### Claim 37 (depends on 34)

The system of claim 34, wherein the delay time is between 8 and 30 hours.

### Claim 38 (depends on 34)

The system of claim 34, wherein the operations further comprise:
receiving a feedback indication concerning the first email message; and
updating, in response to receiving the feedback indication, the negative feedback value for the first email message in the delay buffer to a positive feedback value.

### Claim 39 (depends on 38)

The system of claim 38, where in the operations further comprise:
updating the machine learning model using the positive feedback value; and
constructing and providing a second email message using the updated machine learning model.

### Claim 40

A non-transitory computer readable medium containing instructions that, when executed by at least one processor of a system, cause the system to perform operations comprising:
receiving a feedback indication concerning a first email message sent to a first target, the first email message constructed using a first template and associated with a first context, the feedback indication including an identifier of the first target;
updating, in response to receiving the feedback indication, a feedback value for the first email message;
updating, using the updated feedback data and the identifier, a machine learning model configured to recommend templates based on contexts; and
constructing and providing a second email message using a second template recommended by the updated machine learning model for a second context.
