Query Refinement Engine Using Diversity Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Internet search engines often fail to provide users with the desired search results, as the presented results may be too broad or irrelevant, especially when users submit broad queries or use non-standard terminology, leading to user dissatisfaction.
Innovation Solution
The method involves generating query refinements from user preference data by determining quality and diversity scores for query pairs, associating candidate refinements with initial queries, and filtering them to maintain diversity and relevance, thereby suggesting related queries that may contain the desired results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If search engines present broad search results to respond to user queries, then the quantity of search results is increased, but the relevance and precision of results deteriorates
Solution Approach 1:
The patent segments the search result set into multiple refined subsets based on user preferences and query analysis. Instead of presenting one broad result set, the system divides results into categories or refines the query to generate multiple targeted result sets, thereby maintaining quantity while improving relevance through segmentation.
Solution Approach 2:
The system changes parameters of the search query based on user preferences and behavior data. By dynamically adjusting search parameters (such as relevance thresholds, result types, or filtering criteria), the system can maintain a sufficient quantity of results while improving their relevance to user needs.
2Measurement precision
If search engines use standard terminology and specific queries to improve result precision, then the relevance of search results is improved, but the adaptability to user intent deteriorates
Solution Approach 1:
The patent implements feedback mechanisms that analyze user interactions with search results (clicks, dwell time, refinements) and use this feedback to adapt future search results. This allows the system to maintain precision while becoming more adaptable to individual user intent through continuous learning from user behavior.
Solution Approach 2:
The search system dynamically adjusts its behavior based on real-time user preferences and query context. Rather than using fixed terminology, the system adapts its search strategy, result presentation, and query refinement suggestions to match evolving user intent, thereby maintaining both precision and adaptability.
3Ease of operation
If search engines present query refinements to help users find desired results, then the ease of operation is improved, but the complexity of the search system increases
Solution Approach 1:
The patent enables users to self-serve by providing automated query refinement suggestions based on their own search history and preferences. The system generates and presents refinement options that users can select without requiring complex manual configuration, thereby improving ease of operation while keeping the interface simple despite the underlying system complexity.
Data Source
AI summary
Methods, systems, and apparatus, including computer program products, for generating query refinements from user preference data. A group of query pairs are obtained. Each query pair includes a first query and a second query. A quality score is determined for each query pair from user preference data for documents responsive to both the first and the second query. A diversity score is determined for each query pair having a quality score satisfying a quality threshold, the diversity score determined from user preference data for documents responsive to the second, but not the first, query. For each query pair having a quality score satisfying the quality threshold and a diversity score satisfying a diversity threshold, the second query of the query pair is associated with the first query of the query pair as a candidate refinement for the first query.