Graph Topology Extraction from Distributed Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Extracting and analyzing graph topologies from distributed relational databases is cumbersome and labor-intensive, requiring manual SQL queries, data conversion, and graph algorithms, and existing solutions struggle to detect relationships between multiple databases.
Innovation Solution
A system that extracts graph topology from distributed relational databases using a combination of graph extraction, stitching, and derivation techniques, applying statistical and machine learning methods to create graph descriptions and visualize relationships across multiple data sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual SQL queries and data conversion methods are used to extract graph topologies, then extraction accuracy can be maintained, but the process becomes labor-intensive and time-consuming
Solution Approach 1:
The system performs self-service by automatically extracting graph topologies from distributed relational databases using machine learning models. The automated extraction process eliminates manual SQL query formulation and data conversion, allowing the system to autonomously identify entities, relationships, and graph structures while maintaining high extraction accuracy through trained ML algorithms.
Solution Approach 2:
The patent replaces manual mechanical processes (SQL querying, data conversion scripts, manual graph construction) with an automated machine learning-based system. The ML model directly processes relational database schemas and data to extract graph topologies, substituting the mechanical manual workflow with an intelligent automated system that maintains accuracy while dramatically reducing time consumption.
2Difficulty of detecting and measuring
If existing solutions are used to extract graph topologies, then some relationships can be detected, but the ability to detect relationships between multiple distributed databases is insufficient
Solution Approach 1:
The system achieves universality by designing a machine learning model that can handle multiple types of relational databases and detect various kinds of relationships across different database schemas. The model is trained to recognize entities and relationships in diverse database structures, enabling it to adaptively extract graph topologies from multiple distributed databases with different schemas and relationship types.
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a mediator between distributed relational databases and graph representation. This ML intermediary processes and standardizes data from multiple database sources, transforming heterogeneous relational data into a unified graph topology format, thereby enabling relationship detection across multiple distributed databases that existing solutions cannot handle.
3Productivity
If automated extraction methods are implemented, then processing efficiency improves, but the complexity of the system increases
Solution Approach 1:
The system applies segmentation by dividing the graph extraction process into distinct modular components: database schema analysis, entity identification, relationship detection, and graph construction. Each component is handled by specialized machine learning models or processing modules, which improves extraction efficiency through parallel processing while managing complexity through clear separation of concerns and modular architecture.
Data Source
AI summary
Example embodiments relate to extract graph topology from a plurality of databases. The example disclosed herein access metadata from a plurality of distributed databases. The example further access a set of predetermined rules to transform the accessed metadata into a graph description schema. The example finalizes when the visualization of the graph description schema is built.