Query Classification via Click Log Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engine technologies face challenges in accurately classifying user queries and URLs to provide relevant search results and targeted advertisements, as they often rely on manual generation of keywords or indiscriminate advertising, which can lead to inefficiencies in reaching the right audience.
Innovation Solution
A method involving the analysis of query click logs to determine queries and URLs that correspond to specific categories, where seed documents and queries are used to identify relevant keywords and associate additional queries with categories based on click data, with probabilities determining category assignment within a predefined range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual generation of keywords is used, then keyword accuracy can be maintained, but productivity and efficiency deteriorate
Solution Approach 1:
The system uses automated query classification to generate keywords without requiring manual intervention. The classification system processes query click logs and automatically assigns queries to categories, generating keywords that reflect actual user behavior patterns rather than relying on manual keyword creation.
Solution Approach 2:
The patent replaces manual keyword generation with an automated computational system that analyzes query click logs. The system uses algorithms to classify queries into categories and extract keywords automatically, substituting the mechanical process of manual keyword creation with an automated information processing system.
2Area of stationary object
If indiscriminate advertising is used, then coverage area increases, but relevance and targeting accuracy deteriorate
Solution Approach 1:
The system applies different advertising strategies to different query categories. By classifying queries into specific categories based on user behavior patterns, the system enables localized advertising targeting where ads are displayed only for queries that match the advertisement's category, improving both relevance and targeting accuracy while maintaining appropriate coverage.
Solution Approach 2:
The patent changes the parameter of ad targeting from blanket coverage to category-specific targeting. By using query classification results to determine which ads to display for which queries, the system adjusts the targeting parameters to match user intent, improving ad relevance without significantly reducing overall coverage.
3Measurement precision
If query classification is performed, then search result relevance improves, but device complexity increases
Solution Approach 1:
The classification system segments queries into distinct categories based on user behavior patterns. By dividing the query space into manageable categories, the system can process and classify queries more efficiently, reducing the computational complexity required for accurate classification while maintaining high search result relevance.
Data Source
AI summary
Methods are provided for the classification of search engine queries and associated documents based on search engine query click logs. One or more seed documents or queries are provided that contain content that is representative of a category. A query click log containing information regarding queries entered by at least one user into the search engine and documents subsequently clicked in search engine results corresponding with the queries is analyzed to determine which one or more queries resulted in clicks on the seed documents. Information is stored associating the one or more queries with the category if they resulted in clicks on the seed documents.