Graph Traversal for Cross-Asset Data Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data modeling solutions struggle to effectively model complex relationships across multiple data assets in databases, relying on subject matter expertise that is limited and costly, and entity-relationship models that are specific and isolated, failing to provide comprehensive and reliable cross-asset mergers.
Innovation Solution
The solution maps entity relationships across data assets using graph traversal, integrating various inputs to model complexities and costs, enabling interactive visualization and self-service discovery of relationships with minimal dependency on subject matter expertise, using a graph-based representation to illustrate and navigate cross-asset relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional entity-relationship modeling is used for cross-asset data merging, then subject matter expertise can guide the modeling process, but the solution becomes isolated and specific to particular data assets, reducing comprehensiveness and reliability
Solution Approach 1:
The patent implements a universal graph-based data model that can represent multiple data assets and their relationships in a unified framework. The graph structure allows nodes to represent entities from different data assets and edges to represent relationships between them, enabling comprehensive cross-asset merging while maintaining reliability through consistent relationship modeling across all assets.
Solution Approach 2:
The patent introduces an intermediary graph traversal mechanism that mediates between different data assets. The graph structure acts as an intermediary layer that connects disparate data assets through defined relationships, allowing reliable cross-asset queries and merges without requiring direct integration between each pair of assets.
2Loss of information
If complex queries are performed to merge data across multiple data assets, then meaningful insights can be discovered, but the complexity of modeling relationships between data assets increases significantly
Solution Approach 1:
The patent segments the complex data modeling task into manageable components by representing data assets as nodes and relationships as edges in a graph. This segmentation allows the system to handle complex multi-asset queries by breaking them down into sequential graph traversal steps, reducing the overall modeling complexity while maintaining information integration.
Solution Approach 2:
The graph structure serves as an intermediary that simplifies complex cross-asset queries. Instead of directly modeling complex relationships between multiple data assets, the system uses the graph as an intermediary layer that manages relationships, making the modeling process more manageable and reducing complexity.
3Reliability
If subject matter expertise is extensively used to model cross-asset relationships, then accurate relationships can be established, but the cost and time required for modeling increases
Solution Approach 1:
The patent performs preliminary actions by automatically generating the graph structure and relationships from existing data assets before queries are executed. The system pre-processes data assets to identify entities and relationships, building the graph model in advance, which reduces the time required for actual query execution while maintaining relationship accuracy.
Solution Approach 2:
The system implements self-service capabilities by automatically discovering and modeling relationships between data assets without requiring extensive manual subject matter expertise. The graph-based approach enables the system to self-organize data relationships based on defined schemas and constraints, reducing both time and cost while maintaining reliability.
Data Source
AI summary
There is a need for solutions that perform cross-asset data modeling in a multi-asset database. This need can be addressed by, for example, receiving a request for an execution plan for a merger of a first data entity and a second data entity in the database; determining, based at least in part on a traversal graph of the database, possible paths for the execution plan, wherein each possible path is associated with an ordered combination of path relationships in the database; determining a cost for each possible path based at least in part on at least one of a strength measure associated with each path relationship for the possible path, a traversal cost measure for each path relationship for the possible path, and an experiential usage measure for the possible path; and selecting a recommended path based at least in part on each cost for a possible path.