Document Feature Extraction for Automated Product Question Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating questions and answers about new merchandise require significant manual labor, lacking efficiency and scalability.
Innovation Solution
An information processing apparatus that automatically extracts features from an instruction manual and utilizes accumulated past questions and answers to generate questions about new merchandise without human intervention, by combining related elements and replacing existing merchandise expressions with new ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual extraction of questions from instruction manuals is performed, then question quality and relevance are maintained, but labor cost and time consumption increase significantly
Solution Approach 1:
The system copies the structure and elements from existing Q&A pairs about first target objects to generate questions about second target objects. Instead of manually creating questions from scratch, the system replicates proven question patterns and adapts them to new contexts, significantly reducing manual labor while maintaining question quality
Solution Approach 2:
The system performs preliminary extraction of features from instruction manuals and pre-processes existing Q&A data before question generation. By preparing question templates and element relationships in advance, the system enables automated question generation without requiring manual intervention during the actual question creation process
2Ease of manufacture
If automated question generation is implemented, then manual labor is reduced, but question accuracy and relevance may deteriorate
Solution Approach 1:
The system uses feedback from existing Q&A pairs to improve question generation. By analyzing the structure, wording, and effectiveness of previously generated questions, the system learns to generate more accurate and relevant questions automatically, maintaining high precision while increasing automation
Solution Approach 2:
The question generation system is dynamic and adaptable, adjusting its generation strategy based on the specific features extracted from instruction manuals and the characteristics of existing Q&A data. This dynamic approach allows the system to maintain accuracy across different types of target objects and instruction manual styles
3Loss of information
If features are extracted from instruction manuals and combined with existing Q&A elements, then question relevance improves, but processing complexity increases
Solution Approach 1:
The system segments the question generation process into distinct steps: feature extraction from instruction manuals, element extraction from existing Q&A, and combination of these elements. This segmentation manages complexity by breaking down the complex task into manageable, modular operations that can be processed systematically
Data Source
AI summary
An information processing apparatus includes a processor, the processor extracting, from at least a part of a document, a feature of the part, extracting related elements which are elements related to the feature from a question and an answer about a first target object stored in a storage unit using the extracted feature, and combining the extracted related elements and the feature to generate a question about a second target object specified in the document.