Automated Document Analysis System for Patent Breadth Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current document analysis methods, both manual and automated, face challenges in efficiently analyzing large volumes of documents at scale and speed while maintaining accuracy, especially in tasks requiring subjective judgment, leading to high costs and inconsistencies.
Innovation Solution
An automated system that analyzes documents by filtering, pre-processing, and assigning breadth scores based on word count and commonality, capable of identifying preambles and anomalies, and generating user interfaces to rank document portions, emulating human analysis for industries like law and journalism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis is used to maintain accuracy in subjective judgment tasks, then analysis quality is improved, but cost increases and throughput decreases
Solution Approach 1:
The patent segments the document analysis process into multiple independent rule sets that evaluate different aspects of documents (e.g., claim breadth, novelty, non-obviousness). Each rule set operates autonomously to assess specific criteria, allowing parallel processing of multiple documents simultaneously while maintaining consistent quality standards through structured evaluation frameworks.
Solution Approach 2:
The system implements a universal automated analysis platform that handles multiple document types (patent applications, prior art references, specifications) and multiple evaluation criteria (breadth, novelty, non-obviousness) through a single multi-functional rule-based engine, enabling high-volume processing across diverse analytical tasks without requiring human intervention for each document type.
2Productivity
If multiple different people perform manual analysis to increase throughput, then analysis speed is improved, but inconsistencies increase due to variation in subjective judgment
Solution Approach 1:
The patent applies specialized rule sets tailored to specific evaluation criteria (claim breadth analysis, novelty assessment, non-obviousness determination) rather than using a single generic analysis method. Each rule set contains domain-specific logic and weighting schemes optimized for its particular function, ensuring consistent and reliable evaluation across all documents regardless of which automated process evaluates them.
Solution Approach 2:
The system dynamically adjusts evaluation parameters and weighting factors based on document characteristics and evaluation context. Rule sets modify their assessment criteria and thresholds according to the specific patent area, claim type, and prior art relevance, maintaining consistent quality standards while adapting to varying document complexities and ensuring reliable cross-document comparisons.
3Productivity
If automated document analysis is used to increase throughput and reduce cost, then processing speed is improved, but accuracy in subjective judgment tasks deteriorates
Solution Approach 1:
The patent incorporates feedback mechanisms where rule set evaluations are aggregated and cross-validated against multiple criteria. The system uses feedback loops to adjust rule parameters based on evaluation results, ensuring that automated assessments accurately reflect subjective judgment standards by continuously refining its analysis based on accumulated evaluation data and consistency checks across multiple rule sets.
Solution Approach 2:
The rule-based system acts as an intermediary between raw document data and final analysis conclusions, applying structured evaluation frameworks that translate unstructured patent text into standardized assessments. This intermediary layer processes documents through multiple rule sets that collectively capture the nuances of subjective judgment, achieving high accuracy in automated evaluations by mediating between automated processing and human-like assessment criteria.
Data Source
AI summary
Automatic processing of documents often generates results far different from those obtained by manual human processing. For a given document processing task, many different techniques can be tried but it is often not known which will best emulate manual, human processing. This application discloses data processing equipment and methods specially adapted for a specific application: analysis of the breadth of documents. The processing may include context-dependent pre-processing of documents and sub-portions of the documents. The sub-portions may be analyzed based on word count and commonality of words in the respective sub-portions. Preambles may be identified and analyzed. The equipment and methods disclosed herein improve upon other automated techniques to provide document processing by achieving a result that is quantitatively closer to manual, human processing.