Multi-System Question Answering with Priority-Based Query Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing FAQ systems are limited in their ability to provide answers for a wide range of user questions as they rely on pre-built question and answer pairs, failing to address questions without prior data.
Innovation Solution
A multi-system-based intelligent question answering method that determines question categories, invokes multiple question analysis systems, and processes user queries sequentially through systems with predefined priorities to provide accurate answers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a FAQ system with pre-built question and answer pairs is used, then the system structure is simple and easy to implement, but the system cannot provide answers for questions that are not in the pre-built database, limiting the range of answer needs
Solution Approach 1:
The system is segmented into multiple specialized question analysis systems (FAQ system, knowledge graph system, document analysis system, web search system) instead of using a single monolithic system. Each subsystem handles specific types of questions, allowing the overall system to cover a wide range of answer needs while maintaining simplicity within each component.
Solution Approach 2:
The intelligent question answering system is designed to perform multiple functions by integrating different question analysis systems. The system can handle both simple FAQ-style questions and complex queries requiring knowledge graphs, document analysis, or web search, making it universally applicable to diverse question types without requiring separate systems for each function.
2Measurement precision
If multiple question analysis systems are invoked sequentially, then the accuracy and recall rate improve, but the processing time and system complexity increase
Solution Approach 1:
The system performs preliminary classification of the user's question to determine which question analysis systems are relevant. By pre-categorizing the question type and selecting only the necessary subsystems to invoke, the system avoids unnecessary processing time while maintaining high accuracy through targeted analysis.
Solution Approach 2:
The system invokes question analysis systems partially based on question requirements rather than always using all systems. For simple questions, only the FAQ system is activated; for complex questions, additional systems are invoked. This partial action approach maintains high accuracy when needed while reducing processing time for simpler cases.
3Adaptability or versatility
If multiple question analysis systems are used with system priorities, then the system can handle diverse question types effectively, but the control logic and system management become more complex
Solution Approach 1:
Different parts of the system (different question analysis systems) are assigned different specialties and priorities based on the type of questions they handle best. The FAQ system has high priority for simple questions, while the knowledge graph and document analysis systems have higher priorities for complex queries. This local quality assignment simplifies control logic by creating clear rules for when each system should be invoked.
Data Source
AI summary
The present disclosure provides a multi-system-based intelligent question answering method and apparatus, and a device, relating to the field of artificial intelligence, in particular to the field of knowledge graph. The specific implementation solution is: determining a question category of question information in response to a question answering instruction of a user, wherein the question answering instruction is used to indicate the question information; determining a query engine corresponding to the question category, and invoking multiple question analysis systems corresponding to the query engine according to the query engine; and feeding back answer information to the user when the answer information corresponding to the question information is determined according to a current question analysis system in a process of processing the question information by sequentially using the multiple question analysis systems according to system priorities of the question analysis systems.