TCM Knowledge Graph Construction via NLP Entity Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is currently no knowledge graph in the field of traditional Chinese medicine (TCM) that can provide similar functions to those offered by internet companies' knowledge graphs, such as Google Knowledge Graph or Baidu Knowledge Graph, to manage and present semantic relationships between domain concepts effectively.
Innovation Solution
A method for establishing a TCM knowledge graph by collecting raw data from a TCM database, processing it using natural language processing techniques to obtain structured data, extracting entities and attributes, and constructing the graph using these entities and attributes, which includes relationships like influence, correlation, and dependency between TCM medicine and diseases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional TCM knowledge is organized using conventional databases or text formats, then the data can be stored and accessed, but the semantic relationships between TCM concepts cannot be effectively captured or presented
Solution Approach 1:
The patent transitions from traditional flat database structures to a multi-dimensional knowledge graph structure, adding semantic relationship dimensions (influence, correlation, dependency) to organize TCM concepts. This dimensional transformation enables the capture and presentation of semantic relationships between TCM medicines, diseases, and other concepts that cannot be represented in conventional databases.
2Adaptability or versatility
If TCM knowledge resources are scattered across multiple databases and sources, then comprehensive coverage is achieved, but the knowledge cannot be integrated or accessed systematically
Solution Approach 1:
The patent merges scattered TCM knowledge resources from multiple databases and sources into a unified knowledge graph structure. By integrating data from different TCM databases and organizing them through standardized entities and relationships, the system achieves comprehensive knowledge coverage while enabling systematic access and retrieval.
Solution Approach 2:
The knowledge graph structure serves multiple functions simultaneously: it stores TCM knowledge, captures semantic relationships, enables systematic access, and supports various query types. This multi-functional design allows the same structure to handle diverse TCM knowledge integration and retrieval needs.
3Manufacturing precision
If natural language processing techniques are used to process raw TCM data, then structured data with extracted entities and attributes is obtained, but the processing complexity and computational resources increase
Solution Approach 1:
The patent applies natural language processing techniques in advance to transform raw TCM text data into structured data with extracted entities and attributes before building the knowledge graph. This preliminary processing step organizes unstructured data into a format suitable for knowledge graph construction, improving data structure precision while managing processing complexity through systematic preprocessing.
Data Source
AI summary
A traditional Chinese medicine knowledge graph, a method for establishing a traditional Chinese medicine knowledge graph, and a computer system. The method for establishing the traditional Chinese medicine knowledge graph comprises collecting original data from a traditional Chinese medicine database. The method comprises processing the original data to obtain structural data. The method comprises extracting an entity and an attribute from the structural data. The method comprises constructing the traditional Chinese medicine knowledge graph by utilizing the entity and attribute.