Expert System Query Matching Using Weighted Significance and Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Expert systems face challenges in providing accurate and context-aware natural language query responses, as they struggle to effectively prioritize answers based on the significance of query words and irrelevant content, especially when dealing with diverse knowledge domains and varying user contexts.
Innovation Solution
The implementation of a multi-field expert system that utilizes a search index to evaluate the significance of natural language queries by considering context and relevance, with a private and language-specific storehouse of weighted words, and a process to remove irrelevant words, allowing for the selection of the best matching answers based on weighted scores and context-dependent factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a simple keyword matching approach is used, then the system is easy to operate and fast, but the answer accuracy and context-awareness deteriorate
Solution Approach 1:
The patent segments the query processing into multiple stages: initial candidate generation using simple keyword matching, followed by refinement through significance level evaluation and context analysis. This segmentation allows the system to maintain speed in the first stage while improving accuracy in subsequent stages without overwhelming complexity.
Solution Approach 2:
The patent applies local quality by treating different words in the query differently through significance level assignment. Important words receive higher significance levels and drive the matching process, while less important words receive lower levels. This localized differentiation improves answer accuracy without requiring complete reanalysis of all query components.
2Measurement precision
If all words in the query are treated equally, then the processing is simple and fast, but the significance of key words is lost reducing answer quality
Solution Approach 1:
The patent changes the parameter of word importance by assigning significance levels to different words based on their relevance to the query intent. This parameter transformation allows the system to prioritize key words during matching, improving answer quality without requiring exhaustive analysis of every word, thus maintaining reasonable processing speed.
3Measurement precision
If context information is fully utilized, then the answer relevance improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent performs preliminary action by generating candidate answers using simple keyword matching before applying context analysis. This preliminary stage quickly narrows down the search space, so that subsequent context evaluation only needs to process a limited set of candidates, reducing the overall time penalty while still achieving context-awareness.
Solution Approach 2:
The patent applies partial action by selectively applying context analysis only to candidate answers that pass the initial filtering stage, rather than analyzing all possible answers. This partial application of context evaluation maintains processing efficiency while still achieving sufficient context relevance for the final answer selection.
Data Source
AI summary
The present invention includes an expert system in which a search index furnishes answers to incoming queries provided in natural language. A search index for a specific field contains components that facilitate selecting a best fitting stored answer to the incoming query. Furthermore, context of the incoming query (e.g. location of the user, a current web page or service being used/viewed by the user, the time, etc.) may be considered when selecting a best fitting answer.