Entity-Relationship to Graph Data Mapping for SPARQL Querying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing enterprise resource planning (ERP) systems face difficulties in querying business objects due to the complexity of their native query languages, making it hard to retrieve related objects efficiently, whereas query languages like SPARQL, suited for graph-based data formats, can simplify this process but require data mapping.
Innovation Solution
A method and system that map objects from an entity-relationship model to a graph-based format, allowing them to be queried using a query language like SPARQL, by generating unique identifiers and mapping properties to RDF resources, enabling more intuitive and compact queries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If data is stored in entity-relationship model format in ERP systems, then data structure and storage are simplified, but querying related business objects becomes complex and difficult
Solution Approach 1:
The patent introduces an intermediary layer that maps entity-relationship model data to graph-based data format. This intermediary translation layer allows queries to be executed in the more query-friendly graph format while the underlying storage remains in the compact entity-relationship format, thus resolving the contradiction between storage simplicity and query ease.
Solution Approach 2:
The patent transforms data from a tabular entity-relationship structure into a graph-based structure with nodes and edges, adding a dimensional transformation. This allows related objects to be connected through explicit relationships in the graph, making querying traversals much simpler while maintaining the original storage efficiency.
2Adaptability or versatility
If native query language is used for entity-relationship data, then data format compatibility is maintained, but query complexity increases and efficiency decreases
Solution Approach 1:
The patent employs an intermediary mapping system that translates between entity-relationship format and graph-based format. This allows the system to maintain compatibility with the original data format while enabling efficient graph-based querying, thus improving productivity without sacrificing adaptability.
Solution Approach 2:
The patent creates a copied representation of the entity-relationship data in graph-based format. This copy maintains the essential relationships and structure needed for querying, allowing efficient graph queries to be executed while the original data remains unchanged and compatible.
3Loss of information
If multiple complex queries are used to retrieve related business objects, then complete information can be obtained, but time consumption and operational complexity increase
Solution Approach 1:
By transforming data into graph format with explicit relationship edges, the patent enables single-query traversals to retrieve all related objects across multiple relationship levels. This dimensional transformation eliminates the need for multiple sequential queries, reducing time consumption while maintaining information completeness.
Solution Approach 2:
The graph-based query approach provides a universal method that can retrieve related objects of any type and relationship depth through a single query structure. This multi-functional capability replaces multiple specialized queries, reducing both time and operational complexity while ensuring complete information retrieval.
Data Source
AI summary
Methods and systems that allow of mapping data stored in a first format to a second format is provided. According to the method a data object to be mapped is identified. The object may have a particular type associated with it and one or more associated properties. A unique identifier can be generated for the based on the type of the first object. According to some embodiments, the unique identifier may contain information that can be used to retrieve the object when it is stored in the first format. One or more of the properties associated with the first can be mapped to a resource associated with the unique identifier and also with a second format. The first object can then be queried using a query language associated with the second format.