Dependency-Tree Knowledge Maps for Reliable Document Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI-based techniques struggle to effectively extract knowledge from textual documents, requiring human intervention to differentiate between similar ideas and processes, and are not well-suited for mapping innovation concepts across the product life cycle in manufacturing processes.
Innovation Solution
A system utilizing natural language processing (NLP) models generates dependency trees to organize textual portions into nodes, which are then used to create knowledge maps that visually represent processes, materials, and device configurations, enabling efficient extraction and presentation of manufacturing information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If current AI-based techniques are used to extract knowledge from textual documents, then automation is improved, but accuracy and reliability deteriorate due to inability to differentiate between similar ideas
Solution Approach 1:
The patent segments the knowledge extraction process into multiple specialized modules: dependency tree generation for syntactic analysis, entity extraction for identifying key components, relationship extraction for mapping connections, and knowledge map generation for visual representation. This segmentation allows each module to specialize in specific aspects, improving overall reliability while maintaining automation.
Solution Approach 2:
The patent introduces an intermediary dependency tree structure that serves as a mediator between the input textual document and the final knowledge map. This intermediate representation captures semantic relationships and allows for human review and adjustment, bridging the gap between automated extraction and human expertise to improve reliability.
2Measurement precision
If human expert reading is used to compare documents, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system performs preliminary automated analysis by generating dependency trees and extracting entities and relationships before human review is needed. This preliminary action processes large volumes of text quickly and systematically, reducing the time human experts need to spend on manual comparison while maintaining precision through structured analysis.
Solution Approach 2:
The patent implements feedback mechanisms where the generated knowledge maps and extracted information can be reviewed and corrected by human experts. This feedback loop allows the system to learn from human corrections and improve its automated extraction accuracy over time, balancing speed and precision.
3Quantity of substance
If detailed textual information is extracted, then quantity of information is improved, but device complexity increases
Solution Approach 1:
The patent divides the complex information extraction task into segmented, manageable components: dependency tree generation, entity extraction, relationship extraction, and knowledge map generation. Each component handles a specific aspect of the data, making the overall system more manageable and less complex while still extracting comprehensive information.
Solution Approach 2:
The patent transforms the extracted information from a two-dimensional textual format into a three-dimensional knowledge map structure with nodes, edges, and hierarchical relationships. This dimensional transformation organizes large quantities of information in a visually intuitive manner, making the system more manageable and the information more accessible.
Data Source
AI summary
Systems and methods for extracting information from documents and constructing corresponding knowledge maps with respect to defined knowledge models. Deep-learning-based models for Natural Language Processing (NLP) are applied to tokenize words, tag, parse, and lemmatize sentences of input documents. Then an information extractor traverses the dependency tree of NLP object to recursively extract the entities of interest to the knowledge models. Finally, a knowledge map constructor traverses the dependency tree of NLP object to determine the relationships among the extracted entities and construct knowledge maps recursively following the defined knowledge models.