Clustering System for Dynamic Search Result Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information retrieval systems face challenges in efficiently organizing and refining search results, leading to a discontinuous user experience due to the disconnect between initial and refined search results, and the need for users to manually parse through vast amounts of irrelevant information.
Innovation Solution
A method and system that generates a first set of clusters based on search engine queries, allows for user-driven reclustering of search results, and excludes previously viewed clusters to provide new, relevant information, using linguistic feedback and equivalence classes to refine search results over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If query refinement is implemented by conducting an entirely new search, then relevant information can be obtained, but the search experience becomes discontinuous and confusing for users
Solution Approach 1:
The system dynamically updates the first set of clusters in place to generate the second set of clusters, rather than replacing them entirely. This dynamic updating maintains search continuity while improving relevance, as users can see their original clusters being refined rather than experiencing a complete reset of search results
Solution Approach 2:
The clustering structure is maintained continuously across search refinements, with the second set of clusters being generated by updating the first set rather than replacing it. This preserves the continuous nature of the search experience while allowing relevance improvements through incremental cluster refinements
2Measurement precision
If a narrowly defined search is conducted, then relevant information is obtained, but other valuable information is missed
Solution Approach 1:
The search results are segmented into multiple clusters organized by common properties, topics, and themes. This segmentation allows users to explore different aspects of their search query across multiple clusters, retrieving both highly relevant information and other valuable information that might be missed in a single narrow search result set
Solution Approach 2:
The system adds a dimensional aspect to search results by organizing them into clustered groups with hierarchical relationships. This dimensional organization allows users to navigate through different levels of specificity and explore related information beyond the initial search scope, recovering valuable information that would otherwise be missed
3Adaptability or versatility
If generalized information is searched, then comprehensive coverage is obtained, but users must invest time parsing through irrelevant information
Solution Approach 1:
The system performs preliminary organization of search results into clustered groups based on common properties, topics, and themes before presenting them to users. This preliminary clustering action reduces the time users would otherwise spend parsing through irrelevant information, as results are pre-organized into relevant groups
Solution Approach 2:
The system extracts and separates relevant information from the generalized search results by organizing them into distinct clusters. This extraction process removes irrelevant information from the user's view by grouping it separately, reducing the time needed to parse through comprehensive but unorganized results
Data Source
AI summary
An increase in information available to a user of computing technologies has a tendency to increase the number of topics that are similarly related. Given the large amount of information that is now available, it is increasingly likely that a first set of search results generated in response to an initial search query will contain information that is not of interest to the user. What is needed in the art is a technique to enable a search query to be conducted by taking advantage of linguistic feedback. Furthermore, what is needed is a technique to enable the presentation of search results to be refined in a manner based on what is not of interest to a user, either intrinsically or because the user has already seen and evaluated certain information and next wants to see more or different information.