Question Type-Sensitive Answer Summarization for cQA Reusability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current community-based question and answering (cQA) systems face inefficiencies in reusing answers due to the varying types of questions, where fact-based questions yield reusable answers, while opinion-based and open-ended questions do not, leading to low reusability in information searching processes.
Innovation Solution
A question and answer summarization system identifies question types and generates answer summaries based on categorizations such as reusable vs. not reusable, factual vs. subjective, and unique vs. non-unique answers, employing multi-document summarization techniques to enhance answer reusability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the system stores all answers in the archive without categorization, then the archive size increases and more information is preserved, but the reusability of answers decreases due to inability to efficiently identify applicable answers for different question types
Solution Approach 1:
The patent segments the answer archive by categorizing answers according to question types (fact-based, opinion-based, open-ended) and answer characteristics (reusable, non-reusable, factual, subjective, unique, non-unique). This segmentation enables the system to efficiently retrieve applicable answers by matching question types with categorized answer types, thereby improving answer reusability while maintaining comprehensive archive storage
Solution Approach 2:
The patent applies parameter changes by transforming raw answers into structured answer summaries with specific parameters (question type, answer type, reusability flag). This transformation enables efficient filtering and retrieval based on multiple parameters simultaneously, resolving the contradiction between storing all answers and enabling efficient reuse
2Measurement precision
If the system performs detailed analysis of each question to determine answer reusability, then answer categorization accuracy improves, but the processing time and system complexity increase
Solution Approach 1:
The patent applies preliminary action by pre-categorizing answers during the answer submission process rather than analyzing each question in real-time. Answer summaries are generated and stored with categorization metadata beforehand, enabling fast retrieval without complex real-time analysis, thus maintaining high accuracy while reducing system complexity
Solution Approach 2:
The patent creates simplified copies of answers in the form of answer summaries that contain essential categorization information (question type, answer type, reusability). These summaries serve as lightweight proxies that enable efficient searching and retrieval without requiring complex analysis of the full answer content during query processing
3Loss of information
If the system provides detailed original answers, then the information completeness is maintained, but the retrieval efficiency decreases due to lack of structured formatting for different question types
Solution Approach 1:
The patent segments answers into structured answer summaries with distinct components (question type, answer type, main content, reusability indicator). This segmentation enables the system to quickly identify and retrieve relevant answers by matching query parameters with summary fields, significantly reducing retrieval time while maintaining access to complete original answers when needed
Solution Approach 2:
The patent adds a new dimension of structured metadata to the answer storage system. Answer summaries include categorical dimensions (question type, answer type, reusability) that enable efficient multi-dimensional filtering and retrieval. This dimensional enhancement allows fast search without sacrificing the completeness of original answer content
Data Source
AI summary
In a question answering system, the system identifies a type of question input by a user. The system then generates answer summaries that summarize answers to the input question in a format that is determined based on the type of question asked by the user. The answer summaries are output, in the corresponding format, in answer to the input question.