Graph-Based Natural Language Search System for Patent Novelty Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search tools, including keyword-based and semantic searches, are inefficient for detailed comparisons of technical concepts in large data masses, particularly in patent novelty searches, as they fail to accurately evaluate novelty and account for technical relationships between concepts.
Innovation Solution
A graph-based natural language search system utilizing neural networks to convert documents into data graphs, allowing machine learning algorithms to analyze node values and edges, forming trained models that can accurately determine relevant documents based on technical relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword-based search or semantic search is used, then search speed and document retrieval are improved, but search accuracy and ability to evaluate technical relationships between concepts deteriorate
Solution Approach 1:
The patent segments documents into hierarchical concept graphs where technical concepts are broken down into parent-child relationships. This segmentation allows the system to analyze specific technical relationships between concepts rather than treating documents as monolithic units, thereby improving search accuracy while maintaining efficiency through structured processing.
Solution Approach 2:
The patent transforms traditional flat search into multi-dimensional concept space by creating hierarchical concept graphs with parent-child relationships. This dimensional transformation enables the system to evaluate technical relationships between concepts across multiple levels of abstraction, significantly improving search accuracy for technical queries.
2Measurement precision
If detailed comparison of technical concepts is performed in large data masses, then search accuracy is improved, but computational power requirements and search duration increase
Solution Approach 1:
By segmenting large documents into hierarchical concept graphs, the system processes only relevant concept nodes and their relationships rather than entire documents. This segmentation dramatically reduces computational power requirements while maintaining detailed concept comparison accuracy.
Solution Approach 2:
The system performs preliminary processing to convert documents into structured concept graphs before search operations. This preliminary action organizes technical concepts and their relationships in advance, enabling efficient queries without requiring intensive computational power during actual search operations.
3Adaptability or versatility
If traditional keyword search with boolean strategy is used, then search coverage is improved, but expertise requirement and time consumption increase
Solution Approach 1:
The system automatically constructs hierarchical concept graphs from documents without requiring manual keyword selection or boolean strategy formulation. The automated processing maintains comprehensive search coverage while eliminating the need for user expertise in search query construction, significantly improving ease of operation.
Data Source
AI summary
The invention provides a natural language search system and method. The system comprises a digital data storage means for storing a plurality of blocks of natural language and data graphs corresponding to said blocks. First data processing means are adapted to convert said blocks to said graphs, which are stored in said storage means. The graphs contain a plurality of nodes each containing as node value a natural language unit extracted from said blocks. There are also provided second data processing means for executing a machine learning algorithm capable of travelling said graphs and reading the node values for forming a trained machine learning model based on nodal structures of the graphs and node values of the graphs and third data processing means adapted to read a fresh graph and to utilize said model for determining a subset of said blocks of natural language based on the fresh graph.