# US 12,112,119 B2 · Systems and methods for contextual summarization in an online conversation channel

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

- Number: US 12,112,119 B2
- Document ID: US12112119B2
- Status: Granted
- Kind: B2
- Application number: 17/589,487
- Filed: 2022-01-31
- Published: 2023-09-21
- Granted: 2024-10-08
- Earliest claimed: 2022-01-31
- Assignees: Salesforce, Inc.
- Applicants: Salesforce, Inc.
- Inventors: Sönke Rohde
- Web page: https://patentgrep.com/patent/US12112119B2

## CPC

- G06F 3/04847: Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
- G06F 3/04817: using icons (graphical or visual programming using iconic symbols G06F8/34)
- G06F 16/345: {Summarisation for human users}
- G06F 16/358: Browsing; Visualisation therefor
- G06F 40/106: Display of layout of documents; Previewing
- G06F 40/137: Hierarchical processing, e.g. outlines
- G06F 40/166: Editing, e.g. inserting or deleting
- G06F 40/35: Discourse or dialogue representation
- G06N 3/08: Learning methods
- G06Q 10/107: Computer-aided management of electronic mailing [e-mailing]
- G06Q 10/44: Identification of trends within social networks, e.g. identification of trending topics
- G06Q 10/46: Determination of level of influence of users within social networking services
- G06V 30/10: Character recognition
- H04L 12/1831: {Tracking arrangements for later retrieval, e.g. recording contents, participants activities or behavior, network status}
- H04L 51/216: Handling conversation history, e.g. grouping of messages in sessions or threads
- H04L 51/42: Mailbox-related aspects, e.g. synchronisation of mailboxes
- G06F 40/295: Named entity recognition
- H04L 51/52: for supporting social networking services
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Embodiments described herein provide a contextual conversation summarization mechanism on a chat platform. Specifically, a communication interface receives a first conversation line originated from a first user and a second conversation line originated from a second user. The mechanism then determines that the first conversational line and the second conversational line belong to a same conversation channel based on a channel indicator. The first conversation line and the second conversation line are concatenated as an input to a summarization model. The summarization model then generates a summary of the conversation channel associated with a timestamp. A first visualization element representing a first conversation contribution by the first user and a second visualization element representing a second conversation contribution by the second user are generated based on the generated summary. A user interface visualization of the summary is generated using the first visualization element and the second visualization element.

## Claims

### Claim 1

A method of contextual conversation summarization in a conversation platform, the method comprising:
receiving, via a communication interface, a first conversation line originated from a first user and a second conversation line originated from a second user;
determining that the first conversational line and the second conversational line belong to a same conversation channel based on a channel indicator;
concatenating the first conversation line and the second conversation line as an input to a summarization model;
generating, by the summarization model, a summary of the conversation channel associated with a timestamp;
generating a first visualization element representing a first contribution to the summary that is made by the first user and a second visualization element representing a second contribution to the summary that is made by the second user based on the generated summary, a size of the first visualization element varying with respect to a number of messages generated by only the first user, and a size of the second visualization element varying with respect to a number of messages generated by only the second user;
presenting, via a user interface, the first visualization element and the second visualization element in a visualization structure;
receiving, via the user interface, a user selection of one of the first visualization element or the second visualization element; and
presenting, via the user interface, in response to the user selection, a user interface element extended from the visualization structure that displays at least a part of the summary corresponding to the respective conversation line corresponding to the one of the first visualization element or the second visualization element;
wherein the visualization structure comprises a tree structure;
wherein the first visualization element and the second visualization element are presented as leaf nodes extending from a branch of the tree structure;
wherein branches of the tree structure illustrate the distribution of the first visualization element or the second visualization element over time, and
wherein different branches of the tree structure represent different threads of the conversation.

### Claim 2 (depends on 1)

The method of claim 1, wherein the summary of the conversation is generated by:
identifying a first name of the first user and a second name of the second user from the conversation lines; and
training the summarization model to generate a summary narrative in a format that references the first name or the second user as subjects of sentences.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
extracting a text from an image presented in the first conversation line; and
incorporating the text into the input to the summarization model.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
generating a first summary of the conversation channel based on the input according to a first set of preference parameters for the first user; and
generating a second summary of the conversation channel based on the input according to a second set of preference parameters for the second user,
wherein the first set of preference parameters or the second set of preference parameters are determined based on previous user feedback.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
generating one or more topics of the conversation channel based on the first conversation line and the second conversation line; and
presenting the one or more topics accompanying relevant contributors on a dashboard user interface.

### Claim 6 (depends on 1)

The method of claim 1, wherein the first visualization element further represents a first summary generated relating to the first contribution of the first user to the conversation channel.

### Claim 7 (depends on 1)

The method of claim 1, wherein the generating of the visualization structure further comprises generating a timeline structure having the first visualization element or the second visualization element distributed along a timeline.

### Claim 8 (depends on 1)

The method of claim 1, wherein the generating of the visualization structure further comprises generating a cluster of emojis extracted from the first conversation line and the second conversation line, and wherein the method further comprises:
presenting, in the user interface element extended from the visualization structure, the cluster of emojis.

### Claim 9 (depends on 1)

The method of claim 1 wherein the first visualization element includes a first user icon of the first user and the second visualization element includes a second user icon of the second user.

### Claim 10 (depends on 1)

The method of claim 1, wherein:
the user interface element comprises a sub-element configured to receive a second user selection of the other one of the first visualization element and the second visualization element; and
presenting, via the user interface, in response to the second user selection, the user interface element extended from the visualization structure that displays at least a part of the summary corresponding to the respective conversation line corresponding to the other one of the first visualization element or the second visualization element.

### Claim 11 (depends on 1)

The method of claim 1, wherein the first visualization element and the second visualization element are in a same branch of the visualization structure.

### Claim 12

A system of contextual conversation summarization in a conversation platform, the system comprising:
a communication interface receiving a first conversation line originated from a first user and a second conversation line originated from a second user;
a memory storing a plurality of processor-readable instructions; and
a processor executing the plurality of processor-readable instructions to perform operations comprising:
determining that the first conversational line and the second conversational line belong to a same conversation channel based on a channel indicator;
concatenating the first conversation line and the second conversation line as an input to a summarization model;
generating, by the summarization model, a summary of the conversation channel associated with a timestamp;
generating a first visualization element representing a first contribution to the summary that is made by the first user and a second visualization element representing a second contribution to the summary that is made by the second user based on the generated summary, a size of the first visualization element varying with respect to a number of messages generated by only the first user, and a size of the second visualization element varying with respect to a number of messages generated by only the second user;
presenting, via a user interface, the first visualization element and the second visualization element in a visualization structure;
receiving, via the user interface, a user selection of one of the first visualization element or the second visualization element; and
presenting, via the user interface, in response to the user selection, a user interface element extended from the visualization structure that displays at least a part of the summary corresponding to the respective conversation line corresponding to one of the first visualization element or the second visualization element;
wherein the visualization structure comprises a tree structure;
wherein the first visualization element and the second visualization element are presented as leaf nodes extending from a branch of the tree structure;
wherein branches of the tree structure illustrate the distribution of the first visualization element or the second visualization element over time, and
wherein different branches of the tree structure represent different threads of the conversation.

### Claim 13 (depends on 12)

The system of claim 12, wherein the summary of the conversation is generated by:
identifying a first name of the first user and a second name of the second user from the conversation lines; and
training the summarization model to generate a summary narrative in a format that references the first name or the second user as subjects of sentences.

### Claim 14 (depends on 12)

The system of claim 12, wherein the operations further comprise:
extracting a text from an image presented in the first conversation line; and
incorporating the text into the input to the summarization model.

### Claim 15 (depends on 12)

The system of claim 12, wherein the operations further comprise:
generating a first summary of the conversation channel based on the input according to a first set of preference parameters for the first user; and
generating a second summary of the conversation channel based on the input according to a second set of preference parameters for the second user,
wherein the first set of preference parameters or the second set of preference parameters are determined based on previous user feedback.

### Claim 16 (depends on 12)

The system of claim 12, wherein the operations further comprise:
generating one or more topics of the conversation channel based on the first conversation line and the second conversation line; and
presenting the one or more topics accompanying relevant contributors on a dashboard user interface.

### Claim 17 (depends on 12)

The system of claim 12, wherein the first visualization element further represents a first summary generated relating to the first contribution of the first user to the conversation channel.

### Claim 18 (depends on 12)

The system of claim 12, wherein an operation of generating the visualization structure further comprises generating a cluster of emojis extracted from the first conversation line and the second conversation line, and wherein the operations further comprise: presenting, in the user interface element extended from the visualization structure, the cluster of emojis.

### Claim 19

A processor-readable non-transitory storage medium storing processor-executable instructions of contextual conversation summarization in a conversation platform, the instructions being executed by a processor to perform operations comprising:
receiving, via a communication interface, a first conversation line originated from a first user and a second conversation line originated from a second user;
determining that the first conversational line and the second conversational line belong to a same conversation channel based on a channel indicator;
concatenating the first conversation line and the second conversation line as an input to a summarization model;
generating, by the summarization model, a summary of the conversation channel associated with a timestamp;
generating a first visualization element representing a first contribution to the summary that is made by the first user and a second visualization element representing a second contribution to the summary that is made by the second user based on the generated summary, a size of the first visualization element varying with respect to a number of messages generated by only the first user, and a size of the second visualization element varying with respect to a number of messages generated by only the second user;
presenting, via a user interface, the first visualization element and the second visualization element in a visualization structure;
receiving, via the user interface, a user selection of one of the first visualization element or the second visualization element; and
presenting, via the user interface, in response to the user selection, a user interface element extended from the visualization structure that displays at least a part of the summary corresponding to the respective conversation line corresponding to the one of the first visualization element or the second visualization element;
wherein the visualization structure comprises a tree structure;
wherein the first visualization element and the second visualization element are presented as leaf nodes extending from a branch of the tree structure;
wherein branches of the tree structure illustrate the distribution of the first visualization element or the second visualization element over time, and
wherein different branches of the tree structure represent different threads of the conversation.