AI Model for Non-Explicit Patent Claim Support
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer-generated patent documents lack depth and richness due to the perception that human creativity is essential for providing non-explicit support for claim features, leading to stigma against automated systems.
Innovation Solution
A system and method that utilize machine-readable instructions and machine learning models to parse patent claims into features, align them with corresponding descriptions, identify explicit and non-explicit support, and generate novel text to provide non-explicit support for individual claim features, enhancing the content of computer-generated patent applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computer-generated patent documents are used, then productivity is improved, but the depth and richness of content deteriorates
Solution Approach 1:
The patent introduces an intermediary system comprising natural language processing models and training datasets that bridge the gap between automated generation and human-level content quality. The system uses trained AI models as intermediaries to generate specification text that mimics human creativity and provides non-explicit support for claim features, thereby maintaining both productivity and content depth.
2Loss of information
If human creativity is used to provide non-explicit support, then content depth is improved, but device complexity increases
Solution Approach 1:
The patent replaces the mechanical system of human creativity with an automated AI-based natural language generation system. The system uses trained machine learning models to substitute human intellectual effort, generating non-explicit support text automatically without requiring human intervention, thereby reducing device complexity while maintaining content quality.
3Ease of operation
If automated systems are used, then ease of operation is improved, but content quality deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-training the natural language generation models on extensive datasets of patent specifications and claim-feature support relationships. This preliminary training equips the automated system with the knowledge and patterns needed to generate high-quality content, allowing it to operate autonomously while maintaining specification quality comparable to human-generated content.
Data Source
AI summary
Systems and methods for enhancing the depth and richness of content in computer-generated patent applications by providing non-explicit support for individual claim features are disclosed. Exemplary implementations may: receive a previously unseen claim feature, the previously unseen claim feature being absent from the previously received patent documents; provide one or more sentences of never-been-seen-before computer-generated text using the trained machine learning model and the previously unseen claim feature as input; and insert the one or more sentences of non-explicit support in a draft patent application proximal to explicit support for the previously unseen claim feature.