Intent Resolver Dependency Graph for Multi-Intent Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated computers and machines face challenges in understanding human intentions from user expressions, especially when these expressions contain multiple embedded and dependent intents, making it difficult to respond adequately to human prompts.

Innovation Solution

A system and method for determining user intents by receiving user expressions, extracting entities, gathering associated resolvers, identifying the appropriate resolver based on training expressions, and resolving it to generate an output, utilizing dependency graph resolution to handle multi-intent messages with embedded questions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If automated computers or machines use traditional intent recognition methods, then they can understand simple user expressions, but they fail to adequately respond to expressions containing multiple embedded and dependent intents

Engineering Contradiction:

Improveability to understand multiple embedded and dependent intentsVSAvoidaccuracy of intent determination

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments a complex multi-intent user expression into multiple individual intents, each associated with specific entities and resolvers. The system identifies and processes each intent separately through a structured framework, allowing accurate determination of multiple dependent intents while maintaining reliability in the overall intent recognition process.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If the system processes complex multi-intent expressions by identifying multiple resolvers and their dependencies, then understanding accuracy improves, but processing complexity increases

Engineering Contradiction:

Improveprecision of intent determinationVSAvoidcomplexity of resolver dependency processing

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions by extracting entities from the user expression and gathering associated resolvers before resolving the full intent. This preparatory step organizes the complex information structure in advance, making the subsequent intent determination process more manageable and precise without overwhelming system complexity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary structured framework that mediates between the user expression and the intent determination process. This framework includes entities, resolvers, and dependency relationships that act as intermediaries to systematically process complex multi-intent expressions, improving precision while managing complexity through structured representation.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If the system uses comprehensive entity extraction and resolver gathering methods, then intent recognition accuracy improves, but processing time increases

Engineering Contradiction:

Improveaccuracy of intent recognitionVSAvoidprocessing time for intent determination

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the intent determination process into distinct phases: entity extraction, resolver gathering, and intent resolution. By dividing the comprehensive processing into manageable segments, the system maintains high accuracy through thorough analysis while reducing overall processing time through structured, phased execution rather than monolithic processing.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11481442B2Leveraging intent resolvers to determine multiple intents

Publication Date: 2022.10.25 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

The exemplary embodiments disclose a system and method, a computer program product, and a computer system for determining the intents of user expression. The exemplary embodiments may include receiving a user expression, extracting one or more entities from the user expression, gathering one or more resolvers associated with the one or more entities, identifying a first resolver of the one or more resolvers based on the user expression and the one or more training expressions, and resolving the first resolver to generate a first output.