Dynamic Query Refinement for High Precision Document Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information identification systems fail to adapt to changing user notions of relevance and expand search scope effectively, particularly in complex information needs, leading to missed relevant documents.
Innovation Solution
A method and system that generates and updates queries based on user models and relevance rules, refining them through iterative processes involving user input, document review, and key information extraction to improve accuracy in identifying relevant documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional information identification systems use a fixed relevance definition, then the system operation is simple, but the system cannot adapt to changing user information needs
Solution Approach 1:
The patent implements dynamic query refinement where the system continuously adapts the search query based on user feedback and document review. The relevance definition evolves from a static initial state to a dynamic process that incorporates user interactions, document content analysis, and iterative query updates to match changing user information needs
Solution Approach 2:
The system incorporates feedback loops where user interactions with retrieved documents (reviewing, selecting, rejecting) are fed back into the query refinement process. This feedback mechanism allows the system to learn from user behavior and adjust the search strategy accordingly, improving adaptability without requiring complex manual reconfiguration
2Productivity
If the search scope is limited to highly relevant documents only, then the system operates efficiently, but relevant documents are missed in complex information needs
Solution Approach 1:
The patent applies partial action by initially focusing on highly relevant documents to maintain efficiency, then progressively expanding the search scope as the query evolves. The system performs sufficient but not excessive searching at each stage, adjusting the breadth of search based on the current understanding of user needs and the results obtained so far
Solution Approach 2:
The system performs preliminary search with an initial query to retrieve a subset of highly relevant documents first. This preliminary action provides a foundation for subsequent query refinement and scope expansion, allowing the system to build upon initial results rather than starting from scratch with a broad search
3Measurement precision
If the query is not updated based on document content, then the system operation is simple, but the accuracy of document identification decreases
Solution Approach 1:
The system performs preliminary analysis of document content during the review process to identify key information, concepts, and patterns. This preliminary extraction of meaningful data from documents prepares the foundation for query updates, allowing the system to incorporate actual document insights into the search strategy without requiring complex real-time processing
Solution Approach 2:
The system implements feedback mechanisms where information extracted from reviewed documents feeds back into query refinement. User feedback on document relevance and automated analysis of document content both contribute to updating the query, creating a closed-loop system that continuously improves identification accuracy based on actual evidence from the corpus
Data Source
AI summary
A method and system for performing high precision and high recall relevancy searching is provided. According to embodiments of the present invention, a relevance rule is generated based on a user model and language from within one or more relevant and non-relevant documents. A query is created based on the relevance rule wherein the query may be applied to a corpus to identify relevant and non-relevant documents. The relevance rule may be iteratively refined in order to increase the accuracy of the query. The resulting query may be used by a litigator during the discovery phase of a litigation to respond to a request for production.