Conversational Query Parameterization for Consistent Search Results

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing information output systems struggle to provide consistent search results when conversational sentences dynamically change, despite demanding the same search result, due to variations in wording and phrasing.

Innovation Solution

An information output device and method that includes conversational sentence reception, parameterization, intent determination, search result acquisition, and output, which parameterizes entities in conversational sentences to identify intent and search databases via APIs, ensuring accurate search results even with varying sentence structures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the system uses fixed search queries, then the search result is consistent, but the system cannot handle dynamically changing conversational sentences

Engineering Contradiction:

Improveability to handle dynamically changing conversational sentencesVSAvoidconsistency of search results

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system changes the parameters of search queries by extracting entities and their attributes from conversational sentences. Instead of using fixed queries, the system dynamically adjusts search parameters based on the extracted entities (e.g., employee name, department, time period) to maintain consistent search results across varying sentence structures

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system introduces an intermediary processing layer that includes entity extraction and intent determination modules. This intermediary converts diverse conversational sentences into standardized search parameters, acting as a mediator between the variable user input and the fixed database schema to ensure consistent search results

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If the system processes each conversational sentence individually, then the response is flexible, but the processing complexity increases

Engineering Contradiction:

Improveflexibility in handling user queriesVSAvoidprocessing complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system segments the conversational sentence processing into distinct modules: entity extraction, intent determination, and search query generation. Each module handles a specific aspect of processing, making the overall complex task manageable through modular decomposition while maintaining flexibility in handling diverse user queries

Inventive Principle:

Principle #1Segmentation

3Adaptability or versatility

If the system uses natural language processing, then the system can understand varied sentence structures, but the accuracy of intent recognition decreases

Engineering Contradiction:

Improveunderstanding of varied sentence structuresVSAvoidaccuracy of intent recognition

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system performs preliminary action by pre-defining intent categories and entity types before processing conversational sentences. This preliminary structure guides the natural language processing to accurately recognize intents, improving accuracy while maintaining the ability to handle varied sentence structures through the predefined framework

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11599536B2Information output device, information output method, and information output program

Publication Date: 2023.03.07 SOPPRA CORP

AI summary

An information output receives a conversational sentence, parameterizes one or more entities included in one or more received conversational sentences, searches a database using the to acquire a search result, and outputs the search result.