# US 2025/0156767 A1 · Systems for generating customized datasets

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

- Number: US 2025/0156767 A1
- Document ID: US20250156767A1
- Status: Published application
- Kind: A1
- Application number: 19/026,813
- Filed: 2025-01-17
- Published: 2025-05-15
- Earliest claimed: 2020-08-14
- Assignees: State Farm Mutual Automobile Insurance Company
- Applicants: State Farm Mutual Automobile Insurance Company
- Inventors: Michael Shawn Jacob; Manali B. Desai; Joshua Robert Brown; Oscar Allan Arulfo; Leah Nicole Garcia
- Web page: https://patentgrep.com/patent/US20250156767A1

## CPC

- G06N 20/00: Machine learning
- G06F 16/2228: {Indexing structures}
- G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases}
- G06F 2216/03: Data mining
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

This disclosure provides systems for generating customized datasets. An example dataset generator mines contacts from multiple members of an enterprise, applies a pattern recognition process to determine attributes of the contacts aligned with objectives of the enterprise, stores the contacts and associated attributes in a relational data structure, and through a machine learning process, generates a customized contact dataset of the contacts with attributes relevant to a user request, an event, or a notification received. The example dataset generator may also generate an action steps dataset for the user. The dataset generator may aggregate contacts from calls, chats, emails, SMS messages, and video feeds of an enterprise, apply a pattern recognition process to determine attributes of the contacts, and through a machine learning process, partition and distribute the contacts and associated action steps on respective dashboards of multiple customer service representatives of the enterprise.

## Claims

### Claim 1

A computer-implemented method for determining contact-customized actions, comprising:
executing a data mining engine at an attribute aggregator to acquire a first attribute; and
in response to acquiring the first attribute:
mining contacts from contact datasets;
applying a pattern recognition process to determine a second attribute of the contacts;
determining, by executing a first machine learning process, that the first attribute correlates to the second attribute;
associating, based at least in part on determining that the first attribute correlates to the second attribute, a third attribute with the contacts;
storing the contacts and associated attributes in a first relational data structure;
receiving a user request; and
generating, by executing a second machine learning process:
a customized contact dataset of the contacts associated with the third attribute, and
a customized action steps dataset comprising one or more action steps determined based at least in part on historical actions taken by the contacts prior to generating the customized action steps dataset.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising mining the contacts from one or more of a public database, a social media site, a news source, or an information feed.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein executing the data mining engine to acquire the first attribute comprises generating a previously unrecorded value for the first attribute.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein executing the data mining engine to acquire the first attribute comprises generating a new attribute as the first attribute.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein:
the first attribute is a real property attribute, and
the second attribute is at least one of a vehicle ownership attribute or a pet ownership attribute.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein:
the first attribute is an environmental attribute, and
the second attribute an eligibility attribute.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, further comprising prioritizing the contacts in the customized contact dataset based at least in part on one the user request.

### Claim 8

A system, comprising:
one or more processors; and
non-transitory computer memory storing instructions that, when executed via the one or more processors, cause the one or more processors to determine contact-customized actions by performing operations comprising:
executing a data mining engine at an attribute aggregator to acquire a first attribute; and
in response to acquiring the first attribute:
mining contacts from contact datasets;
applying a pattern recognition process to determine a second attribute of the contacts;
determining, by executing a first machine learning process, that the first attribute correlates to the second attribute;
associating, based at least in part on determining that the first attribute correlates to the second attribute, a third attribute with the contacts;
storing the contacts and associated attributes in a first relational data structure;
receiving a user request; and
generating, by executing a second machine learning process:
a customized contact dataset of the contacts associated with the third attribute, and
a customized action steps dataset comprising one or more action steps determined based at least in part on historical actions taken by the contacts prior to generating the customized action steps dataset.

### Claim 9 (depends on 8)

The system of claim 8, wherein the operations further comprise prioritizing the contacts in the customized contact dataset based at least in part on one the user request.

### Claim 10 (depends on 8)

The system of claim 8, wherein the operations further comprise:
mining strategy elements from user strategy records;
applying the pattern recognition process to determine a fourth attribute of the strategy elements associated with a success objective;
storing the strategy elements and associated attributes in a second relational data structure; and
generating, by executing the second machine learning process, the customized action steps dataset comprising the strategy elements.

### Claim 11 (depends on 10)

The system of claim 10, wherein the operations further comprise:
prioritizing the strategy elements in the customized action steps dataset; and
prioritizing the contacts in the customized contact dataset based on the prioritized strategy elements in the customized action steps dataset.

### Claim 12 (depends on 8)

The system of claim 8, wherein the operations further comprise mining the contacts from one or more of live calls, live chats, current emails, live SMS messages, or live video feeds in real time.

### Claim 13 (depends on 8)

The system of claim 8, wherein the operations further comprise ordering the one or more action steps based at least in part on a queuing order associated with a queuing objective.

### Claim 14

One or more non-transitory computer-readable media storing instructions for determining contact-customized actions that, when executed via one or more processors, cause the one or more processors to perform operations comprising:
executing a data mining engine at an attribute aggregator to acquire a first attribute; and
in response to acquiring the first attribute:
mining contacts from contact datasets;
applying a pattern recognition process to determine a second attribute of the contacts;
executing a first correlation process to determine that the first attribute correlates to the second attribute;
associating, based at least in part on determining that the first attribute correlates to the second attribute, a third attribute with the contacts;
storing the contacts and associated attributes in a first relational data structure;
receiving a user request; and
executing a dataset generator to determine:
a customized contact dataset of the contacts associated with the third attribute, and
a customized action steps dataset comprising one or more action steps determined based at least in part on historical actions taken by the contacts prior to generating the customized action steps dataset.

### Claim 15 (depends on 14)

The one or more non-transitory computer-readable media of claim 14, wherein the operations further comprise partitioning the one or more action steps across a plurality of agents based at least in part on agent attributes.

### Claim 16 (depends on 14)

The one or more non-transitory computer-readable media of claim 14, wherein the operations further comprise executing the dataset generator to:
determine a strategy associated with a user associated with the user request; and
associate the strategy with the customized action steps dataset.

### Claim 17 (depends on 14)

The one or more non-transitory computer-readable media of claim 14, wherein the operations further comprise mining the contacts from one or more of a public database, a social media site, a news source, or an information feed.

### Claim 18 (depends on 14)

The one or more non-transitory computer-readable media of claim 14, wherein the operations further comprise mining the contacts from one or more of live calls, live chats, current emails, live SMS messages, or live video feeds in real time.

### Claim 19 (depends on 14)

The one or more non-transitory computer-readable media of claim 14, wherein executing the data mining engine to acquire the first attribute comprises generating a new attribute as the first attribute.

### Claim 20

A system for determining contact-customized actions, the system comprising:
means for executing a data mining engine at an attribute aggregator to acquire a first attribute; and
in response to acquiring the first attribute, means for:
mining contacts from contact datasets;
applying a pattern recognition process to determine a second attribute of the contacts;
determining, by executing a first machine learning process, that the first attribute correlates to the second attribute;
associating, based at least in part on determining that the first attribute correlates to the second attribute, a third attribute with the contacts;
storing the contacts and associated attributes in a first relational data structure;
receiving a user request; and
generating, by executing a second machine learning process:
a customized contact dataset of the contacts associated with the third attribute, and
a customized action steps dataset comprising one or more action steps determined based at least in part on historical actions taken by the contacts prior to generating the customized action steps dataset.
