Query Rule Generation for Content Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content selection systems struggle to accurately identify and provide relevant content to users' informational needs, especially when search queries do not match targeting keywords, leading to suboptimal user experience and reduced content selection rates.
Innovation Solution
A method involving the segmentation of user session data into subsets based on proximity, classification of search queries into performance groups, and the generation of query rules to identify high-performing queries that can enhance content relevance for low-performing queries, thereby providing additional relevant content to users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional keyword matching is used to select content, then the system is simple and fast, but content relevance is insufficient when search queries do not match targeting keywords
Solution Approach 1:
The patent segments user session data into multiple subsets based on temporal proximity of search queries. Each subset contains queries that are closely related in time, allowing the system to analyze query patterns within each segment. This segmentation enables the system to identify relationships between different queries without overwhelming complexity, improving content relevance while maintaining manageable system architecture.
Solution Approach 2:
The patent performs preliminary classification of search queries into performance groups based on historical content performance data before actual content selection. By pre-analyzing query performance and establishing query rules in advance, the system prepares reference query pairs that link related queries across different performance groups. This preliminary action enables faster, more relevant content selection without adding complex real-time processing during user interactions.
2Productivity
If only the provided search query is used for content selection, then the selection process is simple, but content selection rates are reduced when queries do not match available content
Solution Approach 1:
The patent introduces query rules as intermediary elements that connect search queries to relevant content. These rules are generated from reference query pairs that link queries across different performance groups. When a search query is provided, the system uses the query rule as a mediator to identify additional relevant queries and their associated content, thereby increasing content selection rates without requiring extensive real-time data processing.
Solution Approach 2:
The patent creates reference query pairs by copying and analyzing patterns from high-performing queries to enhance low-performing queries. The system identifies queries that successfully led to content selection and uses these patterns to generate rules that can be applied to similar queries. This copying of successful query patterns enables the system to improve content selection rates by leveraging historical performance data without processing every possible query combination in real-time.
3Reliability
If user session data is analyzed in full detail, then content relevance is improved, but processing complexity and time increase significantly
Solution Approach 1:
The patent divides user session data into multiple subsets based on temporal proximity thresholds. Each subset contains search queries that occurred within a specific time window of each other, making the data more manageable for analysis. This segmentation allows the system to identify query relationships within smaller, more focused groups rather than analyzing the entire user session at once, reducing processing complexity while maintaining content relevance.
Solution Approach 2:
The patent applies partial analysis by focusing only on the most relevant portions of user session data. Instead of analyzing all possible query combinations, the system identifies and processes reference query pairs that meet specific performance criteria. By applying analysis selectively to high-value data segments and using these results to generate query rules, the system achieves good content relevance without the computational burden of exhaustive analysis.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating and applying query rules. Search queries that are received during user sessions are analyzed to generate query rules that specify a high-performing query that can be used to identify content in response to receipt of a low-performing query. The query rules can be generated by identifying queries that are received in a same user sub-session and defining initial query pairs that each have a high-performing query and a low-performing query from the same user sub-session. The initial query pairs that are identified in a threshold number of user sub-sessions are classified as reference query pairs with which query rules are defined. Query rules are made available to a search system, advertisement management system, or another query processing system for identifying content responsive to search queries.