Knowledge Graph Construction for Entity Relationship Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developers face inefficiencies in determining associative relationships between entities in industrial systems due to non-standardized table and column names in user data, relying on personal experience and technical files which are often limited and mismatched.
Innovation Solution
A method and apparatus for constructing a knowledge graph using training data to train a classification model, which identifies entity attributes and mapping relationships between columns, enabling the generation of a knowledge graph that visually represents associative relationships between entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If developers rely on personal experience and technical files to determine associative relationships between entities, then they can identify entity relationships, but the efficiency is low due to limited experience and missing or mismatched technical files
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary structure that stores pre-established relationships between entities. Instead of relying on developers to manually determine relationships from technical files, the system queries the knowledge graph to automatically retrieve entity associations, thereby improving both efficiency and reliability.
Solution Approach 2:
The system performs preliminary actions by pre-building the knowledge graph with entity relationships before the actual data structure identification process. This allows the system to leverage pre-computed entity associations rather than requiring developers to determine relationships on-demand, significantly improving productivity.
2Loss of information
If developers manually determine entities and associative relationships from non-standardized table and column names, then they can understand the data structure, but the process is time-consuming and requires guessing based on limited information
Solution Approach 1:
The patent uses the knowledge graph as a copy or representation of the underlying data relationships. Instead of directly analyzing non-standardized table and column names, the system creates a standardized knowledge graph copy that preserves entity relationships in a structured format, making it easier to extract complete information without time-consuming manual analysis.
Solution Approach 2:
The knowledge graph serves as an intermediary layer between the raw non-standardized data and the developers. It translates ambiguous table and column names into structured entity relationships, preserving information completeness while reducing the time required for interpretation.
3Manufacturing precision
If the system uses standardized entity attributes to represent data columns, then the knowledge graph can accurately represent data structures, but the process of mapping non-standardized column names to entity attributes is complex
Solution Approach 1:
The system employs self-service mechanisms where the knowledge graph automatically performs the mapping between non-standardized column names and standardized entity attributes. The graph structure itself contains the mapping relationships, allowing the system to self-resolve the complexity rather than requiring external manual intervention.
Data Source
AI summary
Various embodiments of the teachings herein include a method for constructing a knowledge graph. The method may include: acquiring training data; using the training data to train a classification model; for each column of data providing each attribute value into the classification model, to obtain an entity attribute, and determining an entity attribute; subjecting to determine a mapping relationship between the columns of data; determining a mapping relationship between the entity attributes; determining a mapping relationship between entities corresponding to the entity attributes; and generating a knowledge graph comprising the entities and the mapping relationship between the entities.