Fault Case Retrieval Using Graded Component Keyword Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
NLP-based fault case retrieval systems often return historical faults that are similar in language description but not closely correlated in actual technology, leading to inaccurate retrieval results.
Innovation Solution
Predefining keywords classified into different grades to determine faults correlated in the actual technology, with higher-grade identical system components indicating higher similarity, and filtering out faults with no identical system components across all grades.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If NLP-based similarity comparison is used to retrieve historical faults, then the retrieval process is simple and fast, but the retrieval accuracy deteriorates because language similarity does not guarantee actual technology correlation
Solution Approach 1:
The patent segments the fault description into structured fields (fault phenomenon, fault cause, fault solution) and extracts keywords from each field. This segmentation allows the system to compare specific technical components rather than performing blanket NLP similarity comparisons, thereby improving accuracy while maintaining efficiency.
Solution Approach 2:
The patent introduces an intermediary keyword extraction and matching mechanism between the NLP-based retrieval system and the actual technology correlation requirement. By extracting standardized keywords from structured fields and comparing these intermediaries, the system bridges the gap between language similarity and technical correlation.
2Measurement precision
If keyword-based technical correlation analysis is performed to improve retrieval accuracy, then the retrieval precision improves, but the processing complexity increases due to multiple analysis steps
Solution Approach 1:
The patent performs preliminary action by pre-structuring fault data into standardized fields (fault phenomenon, fault cause, fault solution) and pre-extracting keywords during data entry. This preliminary structuring eliminates the need for complex real-time analysis, as the keywords and structured fields are already prepared for direct comparison during retrieval.
Solution Approach 2:
The patent changes the parameter representation from continuous NLP similarity scores to discrete keyword matching results. By converting the comparison task into counting identical keywords across structured fields, the system simplifies the processing logic while maintaining high retrieval accuracy.
3Ease of operation
If only language description similarity is considered, then the retrieval process is straightforward, but the retrieved faults are not correlated in actual technology
Solution Approach 1:
The patent applies local quality by focusing comparison on specific local fields (fault phenomenon, fault cause, fault solution) rather than treating the entire text as a uniform block. By extracting and comparing keywords from each localized field separately, the system ensures that technical correlation is assessed in relevant contexts, improving reliability while keeping the process manageable.
Data Source
AI summary
Various embodiments of the teachings herein include a fault processing method comprising: receiving two historical faults similar to a target fault; searching keywords in a description of the target fault and each historical fault, wherein the keywords are classified into N grades, and for each system component in a grade, the grade comprises at least one keyword for describing the component, wherein N is an integer no less than 2; for each of the N grades, counting a quantity of identical system components represented by the keywords in the text description of each historical fault and the target fault; and comparing a degree of similarity of each historical fault to the target fault according to the quantity of identical system components counted in each grade of the N different grades, wherein a historical fault relating to a larger number of high-grade identical system components has a higher degree of similarity to the target fault.