# US 12,217,171 B2 · Engagement signal generation and analysis

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

- Number: US 12,217,171 B2
- Document ID: US12217171B2
- Status: Granted
- Kind: B2
- Application number: 17/245,948
- Filed: 2021-04-30
- Published: 2022-11-03
- Granted: 2025-02-04
- Earliest claimed: 2021-04-30
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Theo Lorrain-Hale; William D. Tierney; Feng Liu; Douglas Lane Milvaney; Manon Knoertzer
- Web page: https://patentgrep.com/patent/US12217171B2

## CPC

- G06N 3/08: Learning methods
- G06Q 10/00: Administration; Management
- G06N 3/04: Architecture, e.g. interconnection topology
- G06F 18/2431: Multiple classes
- G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Engagement signals may be generated and analyzed based on user interactions with documents, particularly in a collaboration environment. The user interactions may generate raw collaboration signals that may be received and processed into cleaned collaboration signals. For example, noise may be removed from the raw collaboration signals to generate the cleaned collaboration signals. The cleaned collaboration signals may be grouped into engagement signals, where each engagement signal represents an individual event or engagement event of the user with the document. The grouping may be based on boundary signals, time frames, and/or any other reasonable limiting element. Each of the engagement signals may be classified into one of several engagement types based on the cleaned collaboration signals in the engagement signal. The engagement signals may then be analyzed to make determinations, recommendations, or the like regarding one or more users of the document, the document content, or the like.

## Claims

### Claim 1

A computer-implemented method for generating and analyzing engagement signals, the method comprising:
for each user of a plurality of users:
receiving raw collaboration signals generated based on the respective user interacting with a document,
processing the raw collaboration signals into cleaned collaboration signals,
grouping the cleaned collaboration signals into engagement signals, each engagement signal representing an individual event, wherein the grouping comprises identifying boundary signals for the individual events with a neural network configured to identify the boundary signals, the identifying comprising:
ingesting, by the neural network, the cleaned collaboration signals; and
outputting, from the neural network, the boundary signals that indicate a beginning and end of the individual events, and
classifying each of the engagement signals into one of a plurality of engagement types based on the cleaned collaboration signals in the respective engagement signal;
generating a collaboration insight comprising collaboration information generated at least in part from the engagement signals of a first user of the plurality of users; and
causing to display, on a display of a computing device of a second user of the plurality of users, a graphical user interface comprising the collaboration insight.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the raw collaboration signals comprise at least one of document open signals, window minimize signals, window focus signals, document close signals, window maximize signals, input activity signals, or a combination thereof.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein processing the raw collaboration signals into the cleaned collaboration signals comprises aggregating the raw collaboration signals to remove noise.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the grouping the cleaned collaboration signals into the engagement signals comprises:
combining the cleaned collaboration signals occurring between the boundary signals into the respective engagement signal.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
analyzing the classified engagement signals of the plurality of users; and
classifying content of the document based at least in part on the analyzing.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the classifying each of the engagement signals into the one of the plurality of engagement types comprises:
providing each of the engagement signals into an intelligence classifier; and
receiving an engagement type as output from the intelligence classifier for each of the engagement signals.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the engagement types comprise editing, viewing, reviewing, and authoring.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, further comprising:
assigning a collaboration role to the first user for the document based on the classified engagement signals.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, further comprising:
providing recommendations to a contributor of the document based on analyzing the classified engagement signals of the plurality of users of the document.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, further comprising:
sending a notification of an action related to the document to one of the plurality of users of the document based at least in part on the classified engagement signals.

### Claim 11

A system comprising:
one or more processors; and
a memory having stored thereon instructions that, upon execution by the one or more processors, cause the one or more processors to:
for each user of a plurality of users:
receive raw collaboration signals generated based on the respective user interacting with a document;
process the raw collaboration signals into cleaned collaboration signals;
group the cleaned collaboration signals into engagement signals, each engagement signal representing an individual event, wherein the instructions to group comprises further instructions to identify boundary signals for the individual events with a neural network configured to identify the boundary signals, the instructions to identify comprising further instructions to:
ingest, by the neural network, the cleaned collaboration signals, and
output, from the neural network, the boundary signals that indicate a beginning and end of the individual events; and
classify each of the engagement signals into one of a plurality of engagement types based on the cleaned collaboration signals in the respective engagement signal,
generate a collaboration insight comprising collaboration information generated at least in part from the engagement signals of a first user of the plurality of users, and
cause to display, on a display of a computing device of a second user of the plurality of users, a graphical user interface comprising the collaboration insight.

### Claim 12 (depends on 11)

The system of claim 11, wherein the raw collaboration signals comprise at least one of document open signals, window minimize signals, window focus signals, document close signals, window maximize signals, input activity signals, or a combination thereof.

### Claim 13 (depends on 11)

The system of claim 11, wherein the instructions to process the raw collaboration signals into the cleaned collaboration signals comprises instructions that, upon execution by the one or more processors, cause the one or more processors to aggregate the raw collaboration signals to remove noise.

### Claim 14 (depends on 11)

The system of claim 11, wherein the instructions to group the cleaned collaboration signals into the engagement signals comprise further instructions that, upon execution by the one or more processors, cause the one or more processors to:
combine the cleaned collaboration signals occurring between the boundary signals into the respective engagement signal.

### Claim 15 (depends on 11)

The system of claim 11, wherein the instructions comprise further instructions that, upon execution by the one or more processors, cause the one or more processors to:
analyze the classified engagement signals of the plurality of users; and
classify content of the document based at least in part on the analysis.

### Claim 16 (depends on 11)

The system of claim 11, wherein the instructions to classify each of the engagement signals into the one of the plurality of engagement types comprises instructions that, upon execution by the one or more processors, cause the one or more processors to:
provide each of the engagement signals into an intelligence classifier; and
receive an engagement type as output from the intelligence classifier for each of the engagement signals.

### Claim 17 (depends on 11)

The system of claim 11, wherein the engagement types comprise editing, viewing, reviewing, and authoring.

### Claim 18 (depends on 11)

The system of claim 11, wherein the instructions comprise further instructions that, upon execution by the one or more processors, cause the one or more processors to:
assign a collaboration role to the first user for the document based on the classified engagement signals.

### Claim 19 (depends on 11)

The system of claim 11, wherein the instructions comprise further instructions that, upon execution by the one or more processors, cause the one or more processors to:
provide recommendations to a contributor of the document based on analyzing the classified engagement signals of the plurality of users of the document.

### Claim 20 (depends on 11)

The system of claim 11, wherein the instructions comprise further instructions that, upon execution by the one or more processors, cause the one or more processors to:
send a notification of an action related to the document to one of the plurality of users of the document based at least in part on the classified engagement signals.