Query Relaxation With Domain Knowledge for Precise Answer Retrieval

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conversational systems face challenges in maintaining high precision and recall when using query relaxation techniques, especially when searching external knowledge sources that are not customized to their domain-specific requirements, leading to undesired information and reduced performance.

Innovation Solution

A query relaxation method that leverages external domain-specific knowledge sources, incorporating offline adaptation to introduce additional relationships between concepts and using a hybrid similarity measure to ensure high precision and recall, by calculating path weights and context-aware information content.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If query relaxation is applied to search external knowledge sources, then query flexibility is improved, but precision and recall deteriorate due to domain-specific vocabulary mismatches

Engineering Contradiction:

Improvequery flexibilityVSAvoidprecision and recall

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary layer (vocabulary mapping module) that translates between domain-specific terminology and general external knowledge source terminology. This mediator enables query relaxation to work effectively by converting domain terms like medical conditions or financial concepts into standardized terms that external knowledge sources can understand, thereby maintaining both query flexibility and precision/recall simultaneously

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically adjusts parameters including vocabulary mapping thresholds, similarity scoring weights, and relaxation depth based on domain-specific requirements. By changing these parameters, the system optimizes the balance between query flexibility and measurement precision for different domains such as healthcare, finance, or technology

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If external knowledge sources are used for query relaxation, then domain vocabulary expansion is achieved, but resource utilization increases

Engineering Contradiction:

Improvedomain vocabularyVSAvoidresource utilization

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by pre-loading and caching vocabulary mappings, domain ontologies, and frequently accessed knowledge graphs into memory before query processing. This advance preparation reduces the computational resources needed during actual query execution, as the heavy lifting of vocabulary expansion has already been done

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements local quality by loading only domain-relevant portions of external knowledge sources rather than entire knowledge bases. The system selectively loads vocabulary and relationships specific to the current domain context (e.g., only healthcare-related terms when processing medical queries), reducing memory usage and processing overhead while maintaining vocabulary expansion benefits

Inventive Principle:

Principle #3Local quality

3Measurement precision

If offline adaptation is applied to incorporate external knowledge sources, then query relaxation performance is improved, but system complexity increases

Engineering Contradiction:

Improvequery relaxation performanceVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the system into distinct modular components: a vocabulary mapping module, an offline adaptation module, a query relaxation module, and a response generation module. Each module has a specific function and can be independently configured, maintained, and optimized. This segmentation reduces system complexity by making the overall system more manageable and easier to understand despite the advanced functionality

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12488012B2Query relaxation using external domain knowledge for query answering

Publication Date: 2025.12.02 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

An offline adaptation method of customizing and incorporating external knowledge domain specific knowledge sources into an existing knowledge base (KB) of a cognitive conversation system, includes obtaining an instance-concept frequency from the existing KB. All the contexts of a domain ontology in which an instance-concept is used are computed. All instance-concepts from the existing KB to an external domain-specific knowledge source are mapped. The existing KB is configured to search for semantically related terms in response to a natural language query received from a conversation system.