# US 12,192,020 B2 · Systems and methods for structuring information in a collaboration environment

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

- Number: US 12,192,020 B2
- Document ID: US12192020B2
- Status: Granted
- Kind: B2
- Application number: 18/480,651
- Filed: 2023-10-04
- Published: 2024-02-01
- Granted: 2025-01-07
- Earliest claimed: 2019-03-10
- Assignees: RingCentral, Inc.
- Applicants: RingCentral, Inc.
- Inventors: Vadim Zhuk; Helen Prask; Ivan Anisimov; William Zhuk
- Web page: https://patentgrep.com/patent/US12192020B2

## CPC

- H04L 12/1822: {Conducting the conference, e.g. admission, detection, selection or grouping of participants, correlating users to one or more conference sessions, prioritising transmission}
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/09: Supervised learning
- G06N 3/096: Transfer learning
- G06N 20/00: Machine learning
- H04L 12/1818: {Conference organisation arrangements, e.g. handling schedules, setting up parameters needed by nodes to attend a conference, booking network resources, notifying involved parties}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
- H04L 51/216: Handling conversation history, e.g. grouping of messages in sessions or threads
- H04L 51/56: Unified messaging, e.g. interactions between e-mail, instant messaging or converged IP messaging [CPM]
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computer-implemented machine learning method for improving a collaboration environment is provided. The method comprises receiving text data for one or more users of the collaboration environment. The method further comprises generating a statement by partitioning the text data. The method further comprises determining an act using the statement and generating a thread using at least the statement and the act. The method further comprises generating an actor list using at least the thread, and generating an actionable item using the actor list and the thread.

## Claims

### Claim 1

A computer-implemented machine learning method for improving a collaboration environment, the method comprising:
receiving text data for one or more users of the collaboration environment;
generating a statement by partitioning the text data;
determining an act associated with the statement using a trained machine learning model;
generating an actor list using at least the statement and the act; and
determining at least one actionable item for an actor from the actor list, wherein the at least one actionable item is sent directly to a display module without storage in a database.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising
generating a thread using at least the statement and the act; and
wherein generating the actor list includes generating the actor list using the at least the statement, the act, and the thread.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, further comprising:
generating a global graph based on at least the statement; and
wherein generating the actor list includes generating the actor list using the at least the statement, the act, and global graph.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein the global graph comprises at least a first and a second groups of nodes connected by lines, wherein the first group of nodes includes at least one of a user identification or a team identification, and the second group of nodes includes at least one message related information, and wherein at least one line connecting a pair of nodes represents the act related to each one of the pair of nodes.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
extracting participant data, named entity data, and conceptual type data from the statement; and
wherein the actor list is indicative of at least one of: the participant data, the named entity data, and the conceptual type data.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein determining the act comprises extracting feature data and semantic vector data from the statement, and processing the feature data and the semantic vector data using the trained machine learning model.

### Claim 7 (depends on 6)

The computer-implemented method of claim 6, wherein extracting feature data and semantic vector data from the statement further comprises:
extracting the feature data and the semantic vector data from the statement using the trained machine learning model.

### Claim 8

A machine learning system for improving a collaboration environment, the system comprising:
a processor;
a memory operatively connected to the processor and storing instructions that, when executed by the processor, cause:
receiving text data for one or more users of the collaboration environment;
generating a statement by partitioning the text data;
determining an act associated with the statement using a trained machine learning model;
generating an actor list using at least the statement and the act; and
determining at least one actionable item for an actor from the actor list, wherein the at least one actionable item is sent directly to a display module without storage in a database.

### Claim 9 (depends on 8)

The machine learning system of claim 8, wherein the instructions further comprising:
generating a thread using at least the statement; and
determining the at least one actionable item based on the thread.

### Claim 10 (depends on 9)

The machine learning system of claim 9, wherein determining the at least one actionable item based on the thread includes determining a first actionable item based on a first question in the thread from a first user, and determining a second actionable item based on a second question in the thread from a second user.

### Claim 11 (depends on 8)

The machine learning system of claim 8, wherein the at least one actionable item includes at least one of: a task, a request, a prioritized message, a to-do list, a check list, a reminder, an alert, a notification, a calendar invite, or a meeting invitation.

### Claim 12 (depends on 8)

The machine learning system of claim 8, wherein the instructions further comprising:
generating an actor representation; and
sending the at least one actionable item to the actor representation.

### Claim 13

A non-transitory, computer-readable medium storing instructions that, when executed by a processor, cause:
receiving text data for one or more users of the collaboration environment;
generating a statement by partitioning the text data;
determining an act associated with the statement using a trained machine learning model;
generating an actor list using at least the statement and the act; and
determining at least one actionable item for an actor from the actor list, wherein the at least one actionable item is sent directly to a display module without storage in a database.

### Claim 14 (depends on 13)

The non-transitory, computer-readable medium of claim 13, wherein the instructions further comprising:
generating a thread using at least the statement and the act; and
wherein generating the actor list includes generating the actor list using the at least the statement, the act, and the thread.

### Claim 15 (depends on 13)

The non-transitory, computer-readable medium of claim 13, wherein the instructions further comprising:
generating a global graph based on at least the statement; and
wherein generating the actor list includes generating the actor list using the at least the statement, the act, and the global graph.

### Claim 16 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the global graph comprises at least a first and a second groups of nodes connected by lines, wherein the first group of nodes includes at least one of a user identification or a team identification, and the second group of nodes includes at least one message related information, and wherein at least one line connecting a pair of nodes represents the act related to each one of the pair of nodes.

### Claim 17 (depends on 13)

The non-transitory, computer-readable medium of claim 13, wherein the instructions further comprising:
extracting participant data, named entity data, and conceptual type data from the statement; and
wherein generating the actor list comprising generating the actor list using the participant data, the named entity data, and the conceptual type data.

### Claim 18 (depends on 13)

The non-transitory, computer-readable medium of claim 13, wherein determining the act comprises extracting feature data and semantic vector data from the statement, and processing the feature data and the semantic vector data using the trained machine learning model.