# US 2023/0419041 A1 · Natural language understanding for creating automation rules for processing communications

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

- Number: US 2023/0419041 A1
- Document ID: US20230419041A1
- Status: Published application
- Kind: A1
- Application number: 17/849,228
- Filed: 2022-06-24
- Published: 2023-12-28
- Earliest claimed: 2022-06-24
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Kuleen MEHTA; Matheus Camasmie PAVAN; Alan THOMAS
- Web page: https://patentgrep.com/patent/US20230419041A1

## CPC

- G06F 40/30: Semantic analysis
- G06F 16/3329: Natural language query formulation
- G06F 16/3344: {using natural language analysis}
- G06F 40/279: Recognition of textual entities
- G06F 40/55: Rule-based translation
- G06Q 10/107: Computer-aided management of electronic mailing [e-mailing]
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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

## Abstract

Methods and systems for generating automation rules based on natural language inputs. In an example, the technology relates to a computer-implemented method for generating automation rules from natural language input. The method includes receiving a natural language input into a communications application for performing an action on communications received by the communications application; providing the natural language input into a trained machine learning model; receiving, as output from the trained machine learning model, a tagged primitive and an identified action from the natural language input; generating an automation rule for performing the action on a subset of communications received by the communications application, the subset of communications corresponding to the tagged primitives; and executing the generated automation rule to perform action on the subset of communications.

## Claims

### Claim 1

A computer-implemented method, the method comprising:
receiving a natural language input, into a communications application, for performing an action on communications received by the communications application;
providing the natural language input to a trained machine learning model;
receiving, as output from the trained machine learning model, a tagged primitive and an identified action from the natural language input;
generating an automation rule for performing the action on a subset of communications received by the communications application, the subset of communications corresponding to the tagged primitives; and
executing the generated automation rule to perform the action on the subset of communications.

### Claim 2 (depends on 1)

The method of claim 1, wherein generating the automation rule comprises:
based on the identified action, identifying a rule template from a plurality of rule templates, the identified rule template having a predefined filter slot for the tagged primitive; and
filling the filter slot of the identified rule template with the tagged primitive.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
causing a display of the generated automation rule;
receiving an edit to the automation rule; and
storing the edited automation rule as a recurrent automation rule.

### Claim 4 (depends on 1)

The method of claim 1, wherein the tagged primitive includes an entity from the natural language input and a primitive tag.

### Claim 5 (depends on 4)

The method of claim 4, wherein the primitive tag is one of a to tag, a from tag, an invite tag, an update tag, a read tag, a flag tag, a date-span tag, or a folder tag.

### Claim 6 (depends on 1)

The method of claim 1, wherein the action is one of a move action, a flag action, a mark-as-read action, a mark-as-unread action, a delete action, or a show action.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
based on a value of the tagged primitive, querying a remote database to resolve the value of the tagged primitive;
receiving a result to the query; and
wherein generating the automation rule includes using the result to the query.

### Claim 8

A system for generating automation rules from natural language input comprising:
a processor; and
memory storing instructions that, when executed by the processor, are configured to cause the system to perform operations comprising:
receive a natural language input to a communications application for performing an action on communications stored by the communications application;
provide the natural language input into a trained machine learning model;
receive, as output from the trained machine learning model, tagged primitives and an identified action from the natural language input;
generate an automation rule for performing the action on a subset of communications received by the communications application, the subset of communications corresponding to the tagged primitives;
execute the generated automation rule to perform the action on the subset of communications;
store the generated automation rule as a recurring rule;
receive a new communication; and
execute the recurring rule to perform the action on the new communication.

### Claim 9 (depends on 8)

The system of claim 8, wherein generating the automation rule comprises:
based on the identified action, identifying a rule template from a plurality of rule templates, the identified rule template having one or more predefined filter slots for the tagged primitives; and
filling the filter slots of the identified rule template with the tagged primitives.

### Claim 10 (depends on 8)

The system of claim 8, wherein the operations further comprise:
cause a display of the generated automation rule
receive an edit to the automation rule; and
store the edited automation rule as a recurrent automation rule.

### Claim 11 (depends on 8)

The system of claim 8, wherein each of the tagged primitives includes an entity from the natural language input and a primitive tag.

### Claim 12 (depends on 11)

The system of claim 11, wherein the primitive tags include at least one of a topic tag, a category tag, a task tag, or an urgency tag.

### Claim 13 (depends on 8)

The system of claim 8, wherein the action is one of a move action, a flag action, a mark-as-read action, a mark-as-unread action, a delete action, or a show action.

### Claim 14 (depends on 8)

The system of claim 8, further comprising:
based on a value of at least one of the tagged primitives, querying a remote database to resolve the value of the tagged primitive;
receiving a result to the query; and
wherein generating the automation rule includes using the result to the query.

### Claim 15

A computer-implemented method for generating automation rules from natural language input, the method comprising:
receiving a natural language input into a communications application for performing an action on communications received by the communications application;
providing the natural language input into a trained machine learning model;
receiving, as output from the trained machine learning model, a tagged primitive and an identified action from the natural language input;
filtering communications based on the tagged primitive to generate a filtered set of communications; and
performing the action on the filtered set of communications.

### Claim 16 (depends on 15)

The method of claim 15, further comprising generating an automation rule for performing the action based on the tagged primitive.

### Claim 17 (depends on 16)

The method of claim 16, wherein generating the automation rule comprises:
based on the identified action, identifying a rule template from a plurality of rule templates, the identified rule template having one or more predefined filter slots.

### Claim 18 (depends on 17)

The method of claim 17, further comprising filling one or more of the filter slots of the identified rule template with the tagged primitive to generate an automation rule.

### Claim 19 (depends on 17)

The method of claim 17, further comprising:
based on a value of the tagged primitive, querying a remote database to resolve the value of the tagged primitive;
receiving a result to the query; and
filling one or more of the filter slots with the received result to the query to generate an automation rule.

### Claim 20 (depends on 19)

The method of claim 19, further comprising:
storing the automation rule as a recurring rule;
receiving a new communication; and
based on the new communication satisfying a value of the one or more filter slots, performing the action on the new communication.
