Patent Mapping System for Reducing Irrelevant Search Results
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current patent search tools produce large numbers of irrelevant results and fail to present results in a manner that allows for quick relevancy determinations, lacking detail on how to adjust searches for relevant results, and present documents in a traditional format.
Innovation Solution
The system employs patent mapping by identifying and structuring patent documents using claim concepts and limitation concepts in a hierarchical fashion, allowing for efficient screening and analysis by interweaving document portions and linking them, and providing user interfaces for quick review and adjustment of search criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If Boolean queries are used to search patent documents, then the search can be performed with simple operators, but large numbers of irrelevant results are produced
Solution Approach 1:
The patent segments the search process into multiple hierarchical levels: (1) initial Boolean query filtering, (2) concept extraction from retrieved documents, (3) concept-based re-ranking and filtering. This segmentation allows the system to first use simple operators for broad retrieval, then progressively refine results through concept analysis, reducing irrelevant results while maintaining operational simplicity.
Solution Approach 2:
The patent introduces concept extraction and concept matching as an intermediary step between the Boolean query and final results. Instead of directly returning all Boolean-matched documents, the system extracts concepts from documents and matches them against search criteria, acting as a mediator that filters and ranks results to reduce irrelevant matches.
2Ease of operation
If traditional patent search tools are used, then documents are presented in traditional format, but quick relevancy determinations cannot be made
Solution Approach 1:
The patent performs preliminary concept extraction and analysis during the search process itself, rather than requiring users to manually review traditional patent documents. Concepts are extracted and matched in advance, and results are pre-ranked based on concept relevance, allowing users to make quick relevancy determinations without manually analyzing entire patent documents.
Solution Approach 2:
The patent employs visual indicators and highlighting (analogous to color changes) to indicate concept matches and relevance levels in search results. This allows users to quickly identify relevant portions of patent documents without reading the entire document, reducing the time required for relevancy determination while maintaining traditional document presentation.
3Adaptability or versatility
If freeform text search is accepted, then user-friendly searching is enabled, but the tools are still limited to Boolean queries
Solution Approach 1:
The patent uses natural language processing and concept extraction as an intermediary layer between the user's freeform text input and the underlying Boolean query system. The system automatically converts freeform text into structured concept searches, maintaining user-friendly input flexibility while managing system complexity through automated processing rather than requiring complex multi-functional search tools.
4Productivity
If search results are not adjusted, then the initial search is completed quickly, but relevant results cannot be identified among irrelevant ones
Solution Approach 1:
The patent performs preliminary concept extraction and relevance scoring during the initial search phase, before final result presentation. This preliminary analysis allows the system to pre-ranking results based on concept match quality, so that while the initial search completes quickly, the most relevant results are positioned prominently for immediate identification.
Solution Approach 2:
The patent implements feedback mechanisms where concept match results inform subsequent search refinements. The system analyzes concept relevance in initial results and uses this feedback to automatically adjust search parameters or suggest refinements, improving result relevancy accuracy while maintaining efficient search completion through iterative rather than exhaustive processing.
Data Source
AI summary
The present subject matter provides systems, methods, software, and data structures for patent mapping, storage, and searching. Some such embodiments include mapping patent documents, claims, and claim limitations. Some further embodiments provide for searching a universe of patent documents by patent document, claim, limitation, class, element, or concept.