Search Query Intent Analysis via Click Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines fail to effectively meet differentiated user demands when the same query keyword is used under different scenarios, as they do not account for varying user intentions and context.
Innovation Solution
A method and apparatus that obtain a search intention from a query keyword by analyzing historical behavior data, including co-occurring query sequences and click entropies, to provide scenario details tailored to specific user needs, which are then output in a designated area of the search result page.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional search engines return results based solely on query keywords, then the search system remains simple and fast, but it fails to meet differentiated user demands under different scenarios
Solution Approach 1:
The system pre-calculates and stores click entropies for historical query keywords before actual search occurs. This preliminary analysis of user behavior data enables the system to quickly determine search intentions during real-time searching without performing complex calculations at query time, thus improving adaptability while controlling complexity
Solution Approach 2:
The patent introduces click entropy as an intermediary metric that mediates between raw user behavior data and search result generation. This intermediary enables the system to infer search intentions without directly analyzing complex user scenarios, achieving differentiated search results while maintaining system simplicity
2Reliability
If the system analyzes historical behavior data to determine search intentions, then user experience improves, but processing time and computational resources increase
Solution Approach 1:
Click entropies are pre-calculated and stored for historical query keywords before actual search operations. This allows the system to quickly retrieve pre-analyzed data during real-time searching, ensuring accurate search intention determination without incurring processing delays
Solution Approach 2:
The system extracts only the essential click entropy metric from comprehensive user behavior data, rather than analyzing all available historical data during searching. This extraction of key features maintains high reliability in search intention accuracy while minimizing processing time and computational resources
Data Source
AI summary
The present disclosure provides a searching method and apparatus, a device and a non-volatile computer storage medium. According to embodiments of the present disclosure, it is possible to obtain a search intention according to the obtained query keyword, and then obtain at least one scenario detail according to the search intention, so that it is possible to output the at least one scenario detail. Since the same search intention might correspond to different scenario demands, directly providing the user with several scenario details corresponding to the search intention can effectively meet differentiated demands when the user uses the same query keyword to search under different scenarios.