Concept Graph Validation for Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing keyword searching techniques are ineffective in locating relevant information within large datasets, as they fail to accurately identify and validate conceptually similar terms across query and document conceptual graphs.
Innovation Solution
A system that validates attributes associated with concept types in query and document conceptual graphs using an onomasticon, attribute knowledge base, and attribute logic engine to identify and verify conceptually similar terms, ensuring their relevance and accuracy before use in search queries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword searching is used to find information in large datasets, then the search process is simple and fast, but the effectiveness of locating relevant information deteriorates
Solution Approach 1:
The patent introduces conceptual graphs as an intermediary structure between keywords and documents. Instead of directly matching keywords to documents, the system transforms keywords into conceptual graphs that represent concepts and their relationships, then matches these graphs against document conceptual graphs. This intermediary layer enables both efficient retrieval and accurate relevance detection by capturing semantic relationships rather than just literal keyword matches.
Solution Approach 2:
The patent changes the search parameters from simple keyword matching to concept type validation using formal concept analysis. The system validates conceptually similar terms against attributes associated with concept types in the query and document conceptual graphs. This parameter transformation from surface-level keyword matching to deep conceptual attribute validation resolves the contradiction between search speed and effectiveness.
2Reliability
If conceptually similar terms are expanded to improve search relevance, then the accuracy of information retrieval improves, but the complexity of validating and processing these terms increases
Solution Approach 1:
The patent performs preliminary action by pre-establishing the attribute knowledge base and onomasticon before the actual search process. Concept types and their associated attributes are defined in advance, allowing the validation process during search to simply check against these pre-computed structures rather than analyzing concepts from scratch. This preliminary preparation reduces the complexity of the validation process while maintaining high retrieval accuracy.
Solution Approach 2:
The system implements feedback through the validation process where conceptually similar terms are tested against the attributes of concept types. Terms that fail to satisfy the attribute requirements are rejected, and only validated terms are used in the search. This feedback mechanism ensures accuracy while the systematic approach to validation manages complexity by providing clear acceptance/criteria for term inclusion.
Data Source
AI summary
According to one embodiment, attributes associated with a concept type of a query conceptual graph are received. A potentially conceptually similar term is received from an onomasticon. The potentially conceptually similar term is validated according to the attributes. According to another embodiment, attributes associated with a concept type of a document conceptual graph are received. A potentially conceptually similar term is received from an onomasticon. The potentially conceptually similar term is validated according to the attributes.