Post-processor Converting Factoid Answers to Structured Relations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deep question answering systems often discard valuable information related to queries after providing an answer, as their diverse and rule-based methods fail to effectively store and reuse relevant factoid answers for future queries.
Innovation Solution
A method and system that converts factoid answers into structured relations for storage in a knowledge base, utilizing answer generators and confidence scores to correlate answers with query terms, allowing for future query retrieval and knowledge base expansion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If factoid answers are discarded after providing an answer to the user, then the system maintains simplicity in operation, but valuable information is lost and cannot be reused for future queries
Solution Approach 1:
The system performs preliminary action by converting factoid answers into structured relations and storing them in a knowledge base before future queries occur. This allows the information to be readily available and reusable when similar queries are asked, eliminating the need to re-process the same information generation steps.
Solution Approach 2:
Instead of permanently discarding factoid answers after use, the system recovers valuable information by extracting and storing it in a structured knowledge base. This enables the recovered information to be reused for future queries, transforming temporary answers into permanent knowledge assets.
2Loss of information
If all generated factoid answers are stored in the knowledge base, then information completeness is improved, but storage space and processing overhead increase significantly
Solution Approach 1:
The system extracts only the essential and valuable information from generated factoid answers, converting them into structured relations that capture the core knowledge. This selective extraction process filters out redundant or less useful information, storing only what is necessary for future query resolution.
Solution Approach 2:
The system changes the parameter representation of factoid answers by transforming unstructured text answers into structured relations with defined schemas. This parameter transformation compresses the information into a more efficient format that reduces storage requirements while maintaining queryability.
3Loss of time
If a structured knowledge base is created to store factoid answers, then query efficiency is improved for repeated queries, but the initial processing time and system complexity increase
Solution Approach 1:
The system performs preliminary action by pre-converting factoid answers into structured relations and storing them in a knowledge base before future queries occur. This upfront processing creates a ready-to-use knowledge repository that enables rapid response to repeated queries, trading initial processing time for long-term query efficiency.
Data Source
AI summary
A method, computer system, and a computer program product for converting a plurality of factoid answers into a plurality of structured relations for storage in a structured knowledge base is provided. The present invention may include receiving a query from a user. The present invention may also include generating a plurality of possible factoid answers. The present invention may then include determining a plurality of confidence scores for the plurality of possible factoid answers. The present invention may then include generating a plurality of certain factoid answers from the plurality of possible factoid answers. The present invention may then include identifying a plurality of special target items. The present invention may further include generating a plurality of structured relations for each certain factoid answer and the identified plurality of special target items. The present include may also include storing the generated plurality of structured relations into the structured knowledge base.