Dynamic Search Result Filtering via User Expectation Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Information retrieval systems often overwhelm users with large search result sets, making it difficult to find relevant information without narrowing down the search query, and existing techniques for improving relevancy and reducing result set size have been inefficient.
Innovation Solution
An information retrieval system that receives a search query and a relevancy expectation score, transforms the score using a polynomial function to determine a volume regulation function, estimates a cut-off score, and generates a filtered list of results based on this score, allowing users to limit the number of search results without altering the query.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the information retrieval system returns all matching documents for a search query, then the completeness of results is improved, but the user is overwhelmed by information overload and cannot efficiently find relevant information
Solution Approach 1:
The patent changes the parameter of result set size from a fixed large number to a dynamically adjusted number based on user expectation scores. The system transforms the user's relevancy expectation score using a polynomial function to determine a volume regulation function, which then estimates a cut-off score to filter the result set size appropriately.
Solution Approach 2:
The patent replaces manual query narrowing with an automated scoring and filtering mechanism. Instead of requiring users to manually refine their queries to reduce result sets, the system automatically applies the volume regulation function and cut-off score estimation to adjust the result set size based on user expectations.
2Quantity of substance
If the user narrows down the search query to reduce the number of results, then the result set size is reduced, but relevant results may be excluded from the results
Solution Approach 1:
The patent incorporates user feedback in the form of relevancy expectation scores to dynamically adjust the result set size. The system uses these scores to transform polynomial functions that determine volume regulation, creating a feedback loop where user expectations directly influence the number of results returned while maintaining relevancy through cut-off score estimation.
3Reliability
If existing techniques for personalized rankings and expertise-based relevancy are applied, then the relevancy of results is improved, but the result set size is not effectively reduced
Solution Approach 1:
The patent makes the result set size dynamic rather than static. Instead of using fixed personalized ranking or expertise-based filtering, the system dynamically adjusts the number of results based on user expectation scores through polynomial transformation and volume regulation functions, allowing the result set size to adapt in real-time to user needs.
Data Source
AI summary
An information retrieval system includes computing hardware, a database operatively coupled with the computing hardware and including documents for delivery to a user in response to a user search as well as a non-transitory, computer-readable memory component operatively coupled with the computing hardware. The non-transitory, computer-readable memory component further includes a receiving module configured to, upon actuation by the computing hardware, receive a search query and a corresponding relevancy expectation score from the user; a search module configured to, upon actuation by the computing hardware, search the database against the search query provided by the user and output an original list of results each of which exhibits a numerical score; and a post-processing module configured to, upon actuation by the computing hardware, perform post-processing of the numerical scores of the results of the original list.