Query Processing via Symbolic Reasoning and Axiom Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in effectively combining information retrieval and symbolic reasoning to generate precise query answers, as they require significant human intervention and resources for capturing domain knowledge and often fail to provide exact matches to queries.
Innovation Solution
A computer-implemented method that combines information retrieval and symbolic reasoning by determining whether a query can be answered using symbolic reasoning, extracting axioms from documents, checking their consistency with a symbolic knowledge base, and generating answers based on consistent axioms, thereby integrating the strengths of both methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If symbolic reasoning is used to answer queries, then precision is improved, but the system cannot handle queries that require domain knowledge not captured in the knowledge base
Solution Approach 1:
The patent introduces an intermediary mechanism that bridges symbolic reasoning and information retrieval. When symbolic reasoning cannot answer a query, the system automatically extracts axioms from unstructured documents and checks their consistency with the knowledge base, serving as a mediator to handle queries beyond the structured knowledge base while maintaining precision through consistency verification.
2Reliability
If information retrieval is used to answer queries, then recall is improved, but precision deteriorates as results do not necessarily match the query
Solution Approach 1:
The system implements a feedback mechanism where information retrieval results are evaluated against the original query requirements. When IR returns results with high recall but low precision, the system uses consistency checking with the symbolic knowledge base to filter and refine results, providing feedback that improves precision while maintaining the high recall advantage of IR.
Solution Approach 2:
The patent uses the symbolic knowledge base as an intermediary to bridge IR and precise answering. Even when using IR for high recall, the system checks extracted axioms against the symbolic knowledge base for consistency, using this intermediary layer to improve precision without sacrificing the recall advantage of information retrieval.
3Loss of information
If manual intervention is used to capture domain knowledge, then knowledge base completeness is improved, but resource consumption increases significantly
Solution Approach 1:
The system implements self-service by automatically extracting axioms from unstructured documents when queries cannot be answered by symbolic reasoning alone. This automated knowledge extraction process eliminates the need for manual domain knowledge capture, reducing human resource consumption while maintaining knowledge base completeness through on-demand axiom extraction and consistency verification.
4Adaptability or versatility
If the system extracts and verifies axioms from documents, then adaptability is improved, but processing time increases
Solution Approach 1:
The system applies partial action by selectively extracting axioms only when symbolic reasoning determines a query cannot be answered from the existing knowledge base. This avoids the excessive processing time of verifying all possible axioms for every query, while still providing adaptability for handling unanswerable queries through targeted axiom extraction and consistency checking.
Data Source
AI summary
Methods, systems and computer program products for query processing are provided herein. A computer-implemented method includes receiving a first query from a user, determining whether the first query is capable of being answered using symbolic reasoning performed on data of a symbolic knowledge base, and executing the symbolic reasoning to generate a first query answer in response to a determination that the first query is capable of being answered using the symbolic reasoning. Axioms are extracted from a plurality of documents when it is determined that a second query is not capable of being answered using the symbolic reasoning. The method further includes determining whether the axioms are consistent with the symbolic knowledge base, and generating a second query answer based on the axioms in response to a determination that the one or more axioms are consistent with the symbolic knowledge base.