Type Projection Query for Instance Space Graph Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing management systems require multiple queries and complex processing to retrieve information about instances based on their relationships, making it inefficient to retrieve hierarchical data sets from databases.
Innovation Solution
The implementation of type projection queries in an instance space graph allows for dynamic generation of queries that retrieve instances and relationships based on specified instance and relationship types, using path notation for filtering, and representing results as hierarchical subgraphs for efficient data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple queries are used to retrieve information about instances based on relationships, then comprehensive information can be obtained, but the retrieval process becomes complex and inefficient
Solution Approach 1:
The patent combines multiple separate database queries into a single unified query operation. By representing instances and their relationships as a graph structure where nodes are instances and edges are relationships, the system can retrieve comprehensive hierarchical information about instances and their related entities in one query, eliminating the need for multiple separate queries and subsequent processing to establish relationships.
Solution Approach 2:
The patent introduces an instance space graph as an intermediary data structure between the database and the query interface. This graph representation serves as a mediator that pre-organizes instances and relationships, allowing complex relationship-based queries to be translated into efficient graph traversal operations rather than multiple database queries with complex join operations.
2Measurement precision
If multiple queries and processing steps are used to establish relationships between instances, then accurate relationship data can be retrieved, but the retrieval time increases
Solution Approach 1:
The patent performs preliminary organization of instance and relationship data into a graph structure before queries are executed. By pre-establishing the graph representation with instances as nodes and relationships as edges in the instance space graph, the system prepares the data in advance for efficient traversal, eliminating the need for time-consuming query execution and relationship establishment during actual data retrieval operations.
Solution Approach 2:
The patent combines the retrieval of instances and their relationships into a single unified operation. Instead of executing separate queries to retrieve instances and then additional queries to retrieve relationships, the system performs a single graph traversal that simultaneously retrieves both instances and their relationships, significantly reducing retrieval time while maintaining complete and accurate relationship data.
3Loss of information
If complex processing is used to retrieve hierarchical data sets, then complete hierarchical information can be obtained, but processing efficiency decreases
Solution Approach 1:
The patent implements dynamic query generation that adapts to the specific hierarchical structure of the instance space graph. The system dynamically determines the traversal depth, node types, and relationship paths based on the query requirements and graph structure, allowing efficient retrieval of complete hierarchical data without unnecessary processing of irrelevant nodes and relationships.
Solution Approach 2:
The patent segments the hierarchical data retrieval into manageable graph traversal operations. By dividing the instance space graph into traversable paths and levels, the system can efficiently retrieve hierarchical information in structured portions rather than attempting to process the entire graph at once, improving processing efficiency while maintaining completeness of hierarchical data.
Data Source
AI summary
Technologies are described herein for applying type projection queries to instance space graphs. Through the utilization of the technologies and concepts presented herein, a type projection may be used to describe the hierarchy of instance types connected by relationships within an instance space graph. A type projection query based on the type projection may be dynamically generated to query instances and relationships within the instance store. Filter criteria for the query can be expressed using a path notation. Results from of the type projection query may be provided as graphs of instances and relationships which are subgraphs of the queried instance space graph. Programmatic traversal of the resultant subgraph can navigate to each instance node within the result space.


