Structured Response Summarization for Communication Sessions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In communication applications, users face difficulties in quickly finding structured responses to questions due to chronological message presentation, leading to inefficient user interactions, increased computational resources usage, and power consumption.

Innovation Solution

A method for detecting structured questions using a machine learning-trained model, generating user interface elements that summarize or present structured responses, reducing user inputs and improving interface organization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If messages are presented chronologically in a list, then the communication session maintains a simple and chronological structure, but the user has to scroll through and scan each message to locate the desired message, which is time-consuming

Engineering Contradiction:

Improveease of locating messageVSAvoidtime to locate message

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent segments the chronological message list by grouping messages according to their associated questions. Instead of a single chronological stream, messages are organized into question-based segments with summary indicators, allowing users to jump directly to relevant question-answer pairs without scanning entire chronological sequences.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary layer between the chronological message list and the user interface. This intermediary automatically generates summary indicators for questions and organizes messages into question-based groups, mediating between the raw chronological data and the user's information needs to enable quick location of desired messages.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If users manually scroll and scan messages to find answers, then no additional processing is required, but the user experience becomes inefficient and bothersome

Engineering Contradiction:

Improveefficiency of finding answersVSAvoiduser interaction simplicity

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent performs preliminary actions by automatically generating summary indicators for questions and organizing messages into question-based groups before the user views them. This pre-processing of information allows users to quickly locate answers without having to manually scan through all messages, significantly improving efficiency while maintaining ease of operation through the intuitive interface.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If the communication application processes and structures messages automatically, then users can quickly find answers, but computational resources and power consumption increase

Engineering Contradiction:

Improvespeed of locating informationVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent applies partial action by selectively generating summary indicators only for messages that contain questions or answers, rather than processing every message in the communication session. This targeted approach provides the productivity benefits of automatic organization while minimizing computational overhead and power consumption by focusing processing only on relevant messages.

Inventive Principle:

Principle #16Partial or excessive action

4Loss of information

If messages are organized by question structure, then users can easily track responses, but the chronological presentation structure is altered

Engineering Contradiction:

Improvecompleteness of communication contextVSAvoidmessage organization structure

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent uses a nested structure where question-based groups are nested within the overall communication session context, and summary indicators are nested within the message groups. This nested organization preserves the chronological context at higher levels while enabling question-based navigation at lower levels, maintaining information completeness without excessive structural complexity.

Inventive Principle:

Principle #7Nested doll (Nesting)

Data Source

PatentUS11586341B2Structured response summarization of electronic messages

Publication Date: 2023.02.21 GOOGLE LLC

AI summary

A system and method for grouping and organizing structured responses in a communication application at a computing device. A structured question in a plurality of messages can be detected based on a structured question model trained via machine learning. A structured question can be a question predicted by the structured question model to have a number of possible answers fewer than a threshold. A user interface element, corresponding to the structured question, can include a structured summarization that includes one or more answers to the structured question present in the plurality of messages from the plurality of users, and/or a structured response template in which at least a subset of possible answers are presented and are selectable. A command to include the generated graphical user interface element in a record of the communication session in a graphical user interface corresponding to the communication application.