Message Thread Identification Using Probability Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conversation-based collaboration tools face challenges in automatically identifying related messages, leading to user confusion and reduced productivity due to the manual marking of threads and commitments, which is cumbersome and time-consuming.

Innovation Solution

A method using a trained message class prediction model to determine the probability of a previous message class resulting in a current message class, extracting previous messages, and assembling them into threads or commitments, with features like Markov inference models and message ranking to automate the process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual marking of threads and commitments is used, then user control and accuracy are improved, but time consumption and operational complexity increase

Engineering Contradiction:

Improveaccuracy of message relationship identificationVSAvoidtime consumption for manual marking

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs automatic message classification and thread identification without requiring user intervention. The natural language analysis engine independently classifies messages and determines relationships, making the system self-sufficient in identifying message threads and commitments.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical marking operations with automated natural language analysis and machine learning algorithms. The system uses trained models to predict message relationships, substituting human cognitive and manual operations with computational processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If manual marking of threads and commitments is used, then user control and accuracy are improved, but operational complexity increases

Engineering Contradiction:

Improveaccuracy of message relationship identificationVSAvoidoperational complexity

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically performs message classification and thread identification without requiring users to manually mark messages or configure complex parameters. The natural language analysis engine independently determines message relationships based on content analysis.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system uses trained models that have learned from example message interactions to automatically replicate the identification of message relationships. The models copy successful classification patterns from training data to automatically classify new messages and identify threads.

Inventive Principle:

Principle #26Copying

3Productivity

If automatic message classification is implemented, then productivity is improved, but system complexity increases

Engineering Contradiction:

Improveefficiency of message organizationVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system divides the complex task of message analysis into distinct classification categories (e.g., commitment, information, question, statement). Each message is independently classified into specific categories, and threads are formed by connecting messages of related types, breaking down the overall complexity into manageable segments.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The natural language analysis engine serves multiple functions: it classifies individual messages, identifies message relationships, determines thread structures, and extracts commitments. This multi-functional approach consolidates what would otherwise require separate systems into a single unified platform.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Productivity

If automatic message classification is implemented, then productivity is improved, but measurement precision may worsen

Engineering Contradiction:

Improveefficiency of message organizationVSAvoidaccuracy of message relationship identification

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary classification of each message into specific categories before establishing relationships between messages. This preliminary sorting enables more accurate subsequent analysis of message connections, as the classification provides a structured foundation for relationship determination.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses trained models that have been refined through feedback from labeled example data. The classification accuracy improves through iterative training where the system learns from correct and incorrect classifications, adjusting its algorithms to enhance precision while maintaining automated processing.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11861463B2Identifying related messages in a natural language interaction

Publication Date: 2024.01.02 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Using a natural language analysis, a current message is classified into a current message class, the current message being a portion of an interaction in narrative text form. Using a trained message class prediction model, a probability of a previous message class having resulted in the current message class is determined. A previous message is extracted from the interaction using the probability, the previous message being a portion of the interaction occurring prior to the current message, the previous message being classified into the previous message class.