Search System Routing Based on Intent and Display Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search systems struggle to accurately understand user search requests, leading to multiple interactions required to obtain desired results, as they fail to effectively combine results from natural language model-based and search engine-based systems.
Innovation Solution
A method that determines a primary and secondary search system based on the user's search intent, using a natural language model-based system for question-type searches and a search engine-based system for resource-type searches, and presents results from these systems with different display attributes to enhance user understanding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single search system is used, then the system complexity is low, but the search accuracy and user satisfaction deteriorate because the system cannot accurately understand diverse search requests
Solution Approach 1:
The search system is segmented into two distinct search systems: a first search system (search engine-based) and a second search system (large model-based). Each system is specialized for different types of search requests, with the first system handling resource-type searches and the second system handling question-type searches. This segmentation allows each subsystem to maintain high accuracy for its specialized domain while avoiding the complexity of a single universal system.
Solution Approach 2:
The system dynamically determines which search system to use based on the search intent. The search intent determination module analyzes the search request in real-time and dynamically selects the appropriate search system (first or second) based on the determined search intent. This dynamic adaptation allows the system to maintain simplicity in operation while achieving high accuracy through context-appropriate system selection.
2Measurement precision
If multiple search systems are used, then the search accuracy improves by addressing different search intents, but the system complexity increases due to managing multiple systems
Solution Approach 1:
The search intent determination module acts as an intermediary that bridges the user's search request and the appropriate search system. It analyzes the search request, determines the search intent (resource-type or question-type), and routes the request to the suitable search system. This intermediary layer simplifies the overall system architecture by providing a clear decision-making mechanism rather than requiring complex integration between multiple search systems.
Solution Approach 2:
The system changes the operational parameter of which search system is active based on the determined search intent. When the search intent is determined to be resource-type, the first search system is activated; when it is question-type, the second search system is activated. This parameter change approach allows the system to switch between different operational modes efficiently, managing complexity through state-based control rather than simultaneous multi-system operation.
3Ease of operation
If the system presents mixed results without distinction, then the information completeness is high, but the ease of operation deteriorates because users cannot quickly identify relevant results
Solution Approach 1:
The search results are presented with differentiated display attributes based on their source and type. Primary search results from the determined search system are displayed with prominent attributes (such as larger font, different color, or positioned at the top), while secondary search results are displayed with secondary attributes. This local quality differentiation allows users to quickly identify and prioritize relevant information while still providing access to comprehensive results, thereby maintaining ease of operation without sacrificing information completeness.
Data Source
AI summary
A method, apparatus, device and medium for searching are provided. In a method, in response to receiving a search request from a user, a search intent of the user is obtained based on the search request. A primary search system and a secondary search system are determined from a natural language model-based search system and a search engine-based search system, respectively, based on the search intent. In a search result for the search request, a primary result matched to the search request from the primary search system and a secondary result matched to the search request from the secondary search system are presented with different display attributes, respectively. Therefore, a search system that is more suitable to the user's search intent may be automatically selected to provide the primary search result. Further, the secondary search result may be used as a supplement to improve the information obtaining efficiency.