Directed Property Graph Search for Targeted Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional database management systems face inefficiencies in executing data searches and lack robust security measures, as they require searching through all columns or vertices for data values, and their homogeneous security architecture makes them vulnerable to attacks.
Innovation Solution
A directed property graph approach is used to store and manage data, where data values are stored as value vertices with associated edges and entity vertices, enabling efficient data searches and additional security layers through encryption and index tables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional database management systems are used to execute data searches, then all columns or vertices must be searched to find data values, but this results in significantly increased search time and reduced search efficiency
Solution Approach 1:
The patent segments the graph database into distinct components: value vertices (storing data values), entity vertices (storing entity information), and edges (representing relationships). This segmentation allows the search to target specific value vertices directly rather than scanning all columns or vertices, dramatically improving search efficiency and reducing search time.
Solution Approach 2:
The patent introduces edges as intermediary elements that connect entity vertices to value vertices. These edges act as indexed pathways that enable direct access to value vertices through entity relationships, eliminating the need to search through all data and significantly reducing search time while maintaining high productivity.
2Reliability
If conventional database management systems with homogeneous security architecture are used, then a single layer of security is provided, but this makes the system vulnerable to attacks
Solution Approach 1:
The patent applies segmentation to security by implementing heterogeneous encryption algorithms across different components of the graph database. Value vertices, entity vertices, and edges can be encrypted with different algorithms and key management strategies, creating multiple security layers that are resistant to attacks targeting a single security approach.
Solution Approach 2:
The patent changes the security parameter from homogeneous to heterogeneous encryption. By using different encryption algorithms (e.g., symmetric, asymmetric, hashing) for different data elements, the system achieves enhanced security robustness and resistance to various attack vectors while maintaining operational reliability.
3Productivity
If all columns or vertices are searched in conventional database systems, then comprehensive data retrieval is achieved, but the device complexity and computational resources required increase significantly
Solution Approach 1:
The patent extracts the search target to specific value vertices that are directly connected to entity vertices through edges. Instead of searching all columns or vertices, the system extracts and queries only the relevant value vertices based on the search criteria, reducing device complexity while maintaining complete data retrieval capability.
Solution Approach 2:
The patent introduces a new dimensional structure to the data model by organizing data into value vertices, entity vertices, and edges. This dimensional change enables targeted searches along specific relationship paths rather than flat comprehensive scans, reducing system complexity while preserving retrieval completeness.
4Adaptability or versatility
If conventional database systems execute information searches with context definition, then all records associated with a data value are retrieved with respect to specific context, but this limits the ability to perform wildcard and similarity searches
Solution Approach 1:
The patent implements dynamic search capability where the search can adapt between context-specific information searches and context-agnostic data searches. The graph structure allows flexible traversal patterns that can incorporate or ignore contextual constraints based on search requirements, enabling wildcard and similarity searches while preserving the ability to perform contextual information searches when needed.
Data Source
AI summary
A system for executing a search in a dataset is provided. The system includes a storage element that stores a directed property graph derived from the dataset. The directed property graph includes entity vertices corresponding to entities of the dataset, edges corresponding to properties of the entities, and value vertices corresponding to data values of the properties. Each edge couples an entity vertex to a value vertex and includes a label indicating an association therebetween. The system further includes processing circuitry that receives a search query including a reference value. The processing circuitry identifies a value vertex having a data value that is associated with the reference value and generates a response to the search query based on labels of edges coupled to the value vertex, and entities of the dataset represented by entity vertices coupled to the value vertex by way of the edges.