Failure Diagnosis Using Integrated Cross-Domain Knowledge Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing failure diagnosis systems struggle to integrate knowledge across various areas due to domain-specific knowledge graphs with different notations and machine learning errors, making it difficult to diagnose failure factors comprehensively in complex systems like automated vehicles.
Innovation Solution
A processor system that integrates multiple knowledge graphs using semantic similarity to generate an integrated graph, which is used to create a diagnostic model through machine learning, enabling accurate failure diagnosis across areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple domain-specific knowledge graphs are used for failure diagnosis, then the comprehensiveness of diagnosis coverage is improved, but the difficulty of integrating knowledge across different areas increases
Solution Approach 1:
The patent merges multiple domain-specific knowledge graphs into a single integrated knowledge graph by evaluating semantic similarity between knowledge items from different domains. This combining approach maintains comprehensive diagnosis coverage while reducing integration complexity through automated similarity-based merging.
Solution Approach 2:
The patent introduces an intermediary mechanism (semantic similarity evaluation) that mediates between different domain-specific knowledge graphs. This intermediary assesses the semantic relationship between knowledge items and facilitates their integration, solving the complexity of direct multi-domain knowledge integration.
2Measurement precision
If knowledge graphs are created based on different domain knowledge, then the specialization and accuracy within each domain is improved, but the semantic compatibility and ease of integration deteriorates
Solution Approach 1:
The patent changes the parameter of knowledge item representation by introducing semantic similarity as a new parameter. This allows domain-specific knowledge graphs to maintain their specialized accuracy while enabling integration through semantic similarity evaluation, transforming the integration process from notation-matching to meaning-based matching.
Solution Approach 2:
The patent creates a universal integration mechanism that works across different domain-specific knowledge graphs. The semantic similarity evaluation serves multiple functions: it preserves domain specificity, enables cross-domain integration, and provides a unified approach for merging heterogeneous knowledge sources.
3Extent of automation
If machine learning algorithms are used to estimate knowledge graphs, then the automation and efficiency of knowledge acquisition is improved, but the occurrence of errors and reduction of knowledge accuracy worsens
Solution Approach 1:
The patent implements feedback through semantic similarity evaluation, which assesses the quality and consistency of knowledge items generated by machine learning. This feedback mechanism identifies and corrects errors in automated knowledge generation, maintaining accuracy while preserving automation benefits.
Solution Approach 2:
The patent performs preliminary semantic similarity evaluation on knowledge items before they are fully integrated into the diagnostic system. This preliminary action filters out erroneous knowledge items generated by machine learning algorithms, preventing error propagation while maintaining efficient automated knowledge acquisition.
Data Source
AI summary
A processor system includes: one or more processors; and one or more memory resources. The memory resource is configured to store a diagnostic program configured to diagnose an electronic system including a mobile body or equipment, and a plurality of knowledge graphs for use in the diagnostic program and including at least a first knowledge graph and a second knowledge graph created based on domain knowledge different from that of the first knowledge graph. The processor is configured to execute the diagnostic program to (1) evaluate a semantic similarity between a first knowledge item included in the first knowledge graph and a second knowledge item included in the second knowledge graph, and (2) generate, when the similarity satisfies a predetermined condition, an integrated graph in which the first knowledge graph and the second knowledge graph are integrated.