Knowledge Map Generation Using Relational Verb Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional knowledge map generation methods struggle with high accuracy and efficiency when handling large amounts of document data, particularly in complex information structures, and require manual processing if no knowledge ontology is available, leading to inefficiencies and increased costs.
Innovation Solution
A method and device for automatically generating a knowledge map by extracting main entities and relational verbs, clustering them into dictionaries, and using a large language model to generate knowledge triplets, with few-shot prompting to handle unclassified relationships, and storing them in a database for efficient knowledge map construction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional manual methods are used to generate knowledge maps, then knowledge maps can be created, but the process requires significant manual effort and time, reducing productivity
Solution Approach 1:
The system automatically extracts entities, relationships, and attributes from documents using NLP technology without requiring manual intervention. The knowledge map generation process serves itself by automatically processing documents, extracting information, and constructing the knowledge map structure, thereby eliminating manual effort and significantly improving productivity
Solution Approach 2:
The patent replaces manual mechanical processing with automated NLP-based information extraction systems. Instead of manual analysis of documents, the system uses computational methods including entity recognition, relationship extraction, and attribute identification to automatically generate knowledge maps, thereby reducing time loss and increasing efficiency
2Measurement precision
If simple algorithms are used for knowledge extraction, then processing is faster, but accuracy and ability to handle complex information structures deteriorate
Solution Approach 1:
The patent segments the knowledge extraction process into distinct modules: entity extraction, relationship extraction, attribute extraction, and knowledge map construction. Each module handles specific aspects of the complex information processing task, allowing the system to maintain high accuracy while managing complexity through structured decomposition of the processing workflow
Solution Approach 2:
The system introduces intermediate processing structures including entity-relationship-attribute triplets and knowledge graphs as mediators between raw document data and the final knowledge map. These intermediaries enable complex information to be processed systematically, maintaining accuracy by breaking down complex relationships into manageable structured representations
3Adaptability or versatility
If knowledge ontology is not available for a domain, then domain flexibility is maintained, but manual ontology creation is required, increasing time loss
Solution Approach 1:
When no knowledge ontology is available for a domain, the system automatically constructs domain-specific ontologies by extracting entities, relationships, and attributes directly from domain documents. This self-service approach eliminates the need for manual ontology creation while maintaining domain adaptability, as the ontology is generated specifically for the target domain's needs
Solution Approach 2:
The system performs preliminary action by automatically building domain ontologies before knowledge map generation when none exist. This preliminary ontology construction prepares the domain structure in advance, enabling subsequent knowledge extraction and map generation to proceed efficiently without manual intervention, thereby reducing time loss while preserving domain versatility
Data Source
AI summary
A method for generating a knowledge map includes: extracting, as main entities, words by applying a key-phrase-extraction language model to the document; extracting a verb representing a relationship between main entities as a relational verb by performing dependency parsing on a sentence including the main entities; clustering the extracted relational verb into clusters to generate a relational verb dictionary; generating a knowledge triplet including a first entity and a second entity among the main entities and a cluster identifier indicating a relationship between the first entity and the second entity based on the relational verb dictionary; and generating the knowledge map based on the knowledge triplet.