Search Intent Recognition for Non-Local O2O Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current O2O search engines are inefficient in handling non-local search queries, requiring users to manually switch cities, leading to complex operations and low efficiency.
Innovation Solution
An information search method that recognizes user intent using feature information and context, employing a non-local search preference determination model to provide both local and non-local search results without the need for city switching, incorporating collapsible display regions and iterative optimization based on user interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If O2O search engines perform search mainly based on local information, then local search results can be provided accurately, but non-local search requirements cannot be satisfied and user operations become complex
Solution Approach 1:
The system dynamically changes the search scope parameter based on user behavior patterns. By analyzing user click history and search context, the system determines whether to perform local-only search or expand to non-local search, thus adapting the search parameters to user needs without requiring manual city switching
Solution Approach 2:
The system automatically performs non-local search based on user behavior analysis without requiring user intervention to switch cities. The search engine itself services the user's non-local search needs by detecting intent patterns and executing appropriate search scopes autonomously
2Adaptability or versatility
If users manually switch city to search for non-local information, then non-local search results can be obtained, but search efficiency decreases and user experience deteriorates
Solution Approach 1:
The system performs preliminary analysis of user search intent by examining search context features and historical click behavior before executing the search. This preliminary action determines the appropriate search scope (local or non-local), eliminating the need for users to manually switch cities and improving search efficiency
Solution Approach 2:
The search engine achieves multi-functionality by handling both local and non-local search requirements through a unified interface. The system adapts its search scope based on user intent recognition, providing versatile search capabilities without requiring separate operations for different search types
3Adaptability or versatility
If the system provides both local and non-local search results, then user search requirements are better satisfied, but search result relevance may decrease
Solution Approach 1:
The system applies different quality standards to different search result portions. Local search results and non-local search results are differentiated and presented with appropriate context, allowing users to identify relevant results based on their actual intent while maintaining overall search result quality
Data Source
AI summary
The present disclosure provides an information search method, apparatus, and system. The information search method includes: a search word sent by a user by using a client is received; search intention of the user is recognized according to feature information of the search word and/or search context feature information; search is performed by using a search policy corresponding to the recognized search intention and according to the search word to acquire information search results associated with the search word; and the information search results are sent to the client, so that the client displays the information search results.