Large Language Model Query Adjustment for Accurate Intent Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional search and recommendation systems struggle with accurately capturing the deep intentions of users due to low matching degrees between search results and user queries, particularly with fuzzy queries, requiring manual adjustments by users.

Innovation Solution

A data query method and apparatus utilizing a large language model to parse user queries, determine object demands based on background information, adjust queries to align with these demands, and perform data queries to obtain accurate results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional search and recommendation systems are used, then users can perform data queries, but the matching degree between search results and user intentions is low

Engineering Contradiction:
Improvematching degreeVSAvoidmanual adjustment requirement
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent introduces an intermediary component (query adjustment module) that acts as a mediator between the user's original query and the search system. This intermediary parses the user intent, adjusts the query parameters, and generates optimized search queries, thereby improving matching degree without requiring direct manual adjustment by users.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements self-service by automatically analyzing user intent and adjusting queries without requiring user intervention. The query adjustment module autonomously performs parsing, parameter optimization, and query refinement, enabling the system to serve itself in improving search accuracy.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If manual query adjustment is required, then users can improve search accuracy, but user time and operational complexity increase

Engineering Contradiction:
Improvequery accuracyVSAvoiduser time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-parsing user queries and pre-adjusting query parameters before the actual search execution. The query adjustment module prepares optimized queries in advance, so when users submit their search requests, the accurate search can be performed immediately without requiring users to spend time on manual adjustments.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If fuzzy queries are processed, then user flexibility is improved, but the ability to capture deep user intention deteriorates

Engineering Contradiction:
Improvequery flexibilityVSAvoidintention capture accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by dynamically adjusting query parameters based on the detected user intent. The system analyzes the semantic meaning of fuzzy queries and modifies parameters such as search depth, result relevance weights, and filtering criteria to capture deep user intentions while maintaining query flexibility.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20250217397A1Data query method and apparatus based on large language model, and computer program product
Publication Date: 2025.07.03 BAIDU USA LLC
  • US20250217397A1 patent drawing
  • US20250217397A1 patent drawing
  • US20250217397A1 patent drawing

AI summary

A data query method and apparatus based on a large language model, an electronic device, a storage medium, and a computer program product are provided. The method may include: parsing a query request of a target object and determining an object demand of the target object, according to object background information of the target object by using a large language model; adjusting the query request according to the object demand, and generating an adjusted query request; and performing data query according to the adjusted query request to obtain a data query result.