User Model Search Result Ranking via Iterative Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing proprietary search engine technologies face challenges in efficiently processing and returning relevant results to users due to large data volumes, proprietary data protection concerns, and user-centric ranking issues, which hinder collaboration and result in irrelevant document presentation.
Innovation Solution
A user-centric approach where a user model is trained by analyzing and ranking result documents without ingesting or storing enterprise data, allowing for personalized and relevant document selection through iterative user feedback, independent of the proprietary search engine or classifier technology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If enterprise copies and stores data from the enterprise document universe outside the enterprise for analysis, then the ability to improve search results is enhanced, but bandwidth consumption, storage requirements, and transfer time increase significantly
Solution Approach 1:
The patent extracts only the necessary result documents from the enterprise document universe for external analysis, rather than copying the entire dataset. This selective extraction approach enables improvement of search results while minimizing bandwidth consumption and storage requirements outside the enterprise.
Solution Approach 2:
The patent segments the large enterprise document universe into manageable result document sets that can be analyzed externally. By dividing the data into smaller, targeted portions for analysis and feedback, the system reduces the overall data volume that needs to be transferred and stored outside the enterprise.
2Adaptability or versatility
If enterprise exposes proprietary data outside the enterprise for search analysis, then collaboration and result improvement are enhanced, but data security and proprietary information protection are compromised
Solution Approach 1:
The patent introduces an intermediary analysis system that operates outside the enterprise to evaluate search results and provide feedback, without requiring direct exposure of proprietary enterprise data. This intermediary layer enables collaboration and improvement while maintaining data security boundaries.
Solution Approach 2:
The patent creates a copy of only the necessary result documents for external analysis, rather than exposing the entire proprietary enterprise document universe. This selective copying approach enables external collaboration while minimizing the exposure of sensitive proprietary information.
3Reliability
If enterprise uses internal business model ranking for result documents, then enterprise interests are protected, but user-centric relevance and satisfaction are reduced
Solution Approach 1:
The patent implements a feedback mechanism where user interactions with result documents are analyzed to generate improvement signals. This feedback loop enables the system to balance enterprise business model requirements with user-centric relevance by continuously learning from user behavior patterns.
Solution Approach 2:
The patent introduces dynamic adjustment capabilities that allow the ranking system to adapt between enterprise business model constraints and user relevance requirements. The system dynamically modifies result presentation based on user feedback while maintaining enterprise policy compliance.
Data Source
AI summary
Systems and methods for improving search results from proprietary search engine technologies and proprietary machine classifiers, without ingesting, copying, or storing, the data to be searched, are described herein. A user sends a query to a proprietary search engines and gets a result document set back. The user may apply a user model for classifying a result document set to generate a result document for review of a user. The reviewed document may be added to a user training corpus, which is then used to retrain the user model. The retrained user model may be applied by the user to generate the next result document for user review and so on until the user model converges to generate relevant documents reliably. Once the user model converges, the user may apply the now reliable user model to generate multiple relevant documents for the user.