Conceptual Graph Referent Determination for Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing keyword searching techniques fail to find conceptually similar terms for concepts, making it difficult to retrieve relevant information from large datasets.
Innovation Solution
A method for determining query return referents in conceptual graphs by generating a conceptual graph, identifying conceptually similar terms, and associating referents with graph terms needing referents, using a system comprising a conceptual graph generator, term expander, onomasticon manager, graph matcher, and context generator.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If keyword searching is used to find information in a corpus, then the search process is simple and fast, but conceptually similar terms for concepts cannot be found
Solution Approach 1:
The patent introduces conceptual graphs as an intermediary layer between keywords and information. The system generates conceptual graphs from documents and queries, where nodes represent concepts and edges represent relationships. This intermediary structure enables the system to understand and match conceptually similar terms even when exact keyword matches are absent, thereby improving retrieval accuracy without requiring complex semantic analysis of every document.
Solution Approach 2:
The system performs preliminary action by pre-generating conceptual graphs for documents and storing them in a database. When a query arrives, the system generates a conceptual graph for the query and compares it with pre-generated document graphs. This approach avoids the need for complex real-time semantic analysis, maintaining system simplicity while improving retrieval accuracy through pre-processed conceptual representations.
2Adaptability or versatility
If conceptual graphs with concept types are used instead of simple keyword searching, then conceptually similar terms can be found, but the system complexity increases
Solution Approach 1:
The patent segments the search system into distinct functional modules: a graph generator that creates conceptual graphs from text, a graph database for storage, and a graph matcher for comparison. Each module handles a specific aspect of the conceptual graph processing. This segmentation allows the system to implement complex conceptual matching while maintaining manageable system architecture through clear separation of concerns and modular design.
3Reliability
If referent data is determined and stored for future queries, then future search accuracy is improved, but storage requirements and processing time increase
Solution Approach 1:
The system performs preliminary action by pre-determining referent data for concept types in conceptual graphs and storing this information in the graph database. When concept nodes are created during graph generation, the system automatically queries the corpus for referent instances and stores these associations. This pre-processing approach ensures that future queries can leverage pre-computed referent information, improving accuracy while minimizing real-time processing requirements.
Data Source
AI summary
In one embodiment, a method for determining query return referent data for concept types in a conceptual graph includes generating a conceptual graph for a document query, the conceptual graph comprising a plurality of graph terms, identifying one of more of the plurality of graph terms needing referent data, identifying conceptually similar terms for each of the graph terms, for the graph terms needing referents, identifying referent data by searching for instances where conceptually similar terms for graph terms needing referent data are associated by conceptually similar terms for the linking concept term, and associating identified referent data with the graph terms needing referent data.