Question Generation Device for Ambiguous Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine comprehension-type question answering techniques face challenges in achieving high answer accuracy due to ambiguous or short question contents, leading to potential errors in determining unique answers.
Innovation Solution
A question generation device that uses a machine learning model to generate revised questions by supplementing potentially defective portions of input queries with words from a prescribed lexical set, based on relevant documents, to enhance question clarity and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a machine comprehension-type question answering technique is used to automatically answer user questions, then the system can provide automated responses, but the answer accuracy deteriorates when question contents are ambiguous or too short
Solution Approach 1:
The system performs preliminary action by generating a revised question before attempting to answer it. The question generation unit creates a more complete version of the input question by supplementing potentially defective portions with words from the relevant document, ensuring the question contains sufficient information for accurate answering.
Solution Approach 2:
The patent introduces an intermediary element - the revised question - between the original ambiguous question and the answer generation process. This intermediary serves as a mediator that bridges the gap by incorporating necessary information from the relevant document while maintaining the original question's intent.
2Ease of operation
If the question content is kept simple and short for ease of user input, then the ease of operation is improved, but the information completeness deteriorates leading to ambiguous queries
Solution Approach 1:
The system performs preliminary action by generating a revised question before attempting to answer it. The question generation unit creates a more complete version of the input question by supplementing potentially defective portions with words from the relevant document, ensuring the question contains sufficient information for accurate answering.
Solution Approach 2:
The system applies self-service by automatically enriching the user's simple query without requiring the user to manually expand it. The question generation unit autonomously identifies missing information and supplements the query using words from the relevant document, allowing the system to serve itself in completing the information gap.
3Measurement precision
If the question content is expanded to include more information for clarity, then the measurement precision of the query is improved, but the device complexity increases
Solution Approach 1:
The system performs preliminary action by generating a revised question before attempting to answer it. The question generation unit creates a more complete version of the input question by supplementing potentially defective portions with words from the relevant document, ensuring the question contains sufficient information for accurate answering.
Solution Approach 2:
The question generation unit serves multiple functions: it identifies defective portions of the question, selects appropriate words from the relevant document, and generates a revised question. This multi-functional component handles various aspects of query refinement without requiring separate specialized modules for each task.
Data Source
AI summary
A question generation device includes: generating means which uses a query and a relevant document including an answer to the query as input and, using a machine learning model having been learned in advance, generates a revised query in which a potentially defective portion of the query is supplemented with a word included in a prescribed lexical set.