Accessing relational databases as resource description framework databases

• US 8,719,252 B2
• Filed: 10/21/2011
• Issued: 05/06/2014
• Est. Priority Date: 10/22/2010
• Status: Active Grant

• Current Assignee: Data.World, Inc.
• Original Assignee: Daniel P. Miranker, Juan F. Sequeda
• Inventors: Miranker, Daniel Paul, Sequeda, Juan Federico
• Primary Examiner(s): GIRMA, ANTENEH B
• Application Number:
  US13/278,907
• Publication Number:
  US 20120102022A1
• Time in Patent Office: 928 Days
• Field of Search: None
• US Class Current: 707/713
• CPC Class Codes: G06F 16/2452 Query translation
• CPC Class Codes: G06F 16/258 Data format conversion from or to a database

Abstract

This invention is a system and method for integrating relational databases into a semantic web framework utilizing a simple mapping process and the SQL query optimizer present in the SQL database engine.

Claims

1. A method executed by a computer system of executing a digitally encoded database query comprised of a representation expressed in a first query language where the first query language operates on a first type of data organized as a graph, where the query is executed on a second type of data organized as a relational database with a corresponding database schema, by using the computer system to automatically translate the query representation expressed in the first query language into a digitally encoded query representation expressed in a second query language that operates on the relational database by using a subset of the contents of a data structure representing a mapping from the relational database schema to a synthetic domain model that is a putative ontology automatically created from the relational schema.

2. The method of claim 1 further comprising executing the translated query using a computer program that takes queries as input and returns as output a subset of data from a larger set of data.

3. The method of claim 1 where the first query language is SPARQL, the first type of data organization is a semantic web representation, and the second query language is SQL and the computer system is comprised of a SQL database.

10. The methods of claim 1, 2-4, 5-6, or 7-9 with the additional step of optimizing the translated query.

11. The method of claim 10 where the optimization step is comprised of detection of unsatisfiable conditions.

12. The method of claim 10 where the optimization step is comprised of self-join elimination.

13. The method of claim 10 where the optimization step is comprised of self outer join elimination.

14. A system comprised of one or more central processing units operatively connected to computer memory containing program code data that when executed by the one or more central processing units, causes the one or more central processing units to execute the methods of any of the claim 1, 2-4, 5-6, or 7-9.

15. A computer readable data storage device comprising program data, that when executed, causes one or more central processing units to execute any one of the methods of claim 1, 2-4, 5-6, or 7-9.

19. The method of either claims 1, 4 or 7 further comprising:

creating an RDF logical representation using a view definition associated with the relational database.

20. The method of claim 19 where the view definition is a SQL view definition.

21. The method of claim 19 where the view definition is logical.

22. The method of claim 19 where the view definition is materialized.

23. The method of claim 20 where the body of the SQL view definition is nested in the translated query.

24. The method of claim 19 where the translating step is further comprised of outputting a relational query where the arguments of the relational query are the columns of the RDF logical representation of the relational database data contents.

25. The method of either claim 1, 4, 5 or 7 where the translating step is further comprised of converting a URI into a string comprised of a combination of a table name identifying a data table and a column label associated with the identified table.

30. The method of claim 14 with the additional step of optimizing the translated query.

31. The method of claim 15 with the additional step of optimizing the translated query.

4. A method executed by a computer system for the execution of a digitally encoded first query expressed in a semantic query language on the contents of a pre-existing relational database with a corresponding relational database schema, said method comprising

automatically generating a data structure representing a mapping of the pre-existing relational database schema from the relational database schema to a synthetic domain model that is a putative ontology automatically created from the relational schema;

and

automatically translating the first query into a second digitally encoded relational query by using a subset of the contents of the mapping data structure to determine component substitutions of at least one of the components comprising the first query in order to generate the second relational query translation.

16. The method of claim 4 where the synthetic domain model is expressed in OWL.

17. The method of claim 4 where the synthetic domain model is expressed in RDFS.

18. The method of either claims 4, 5 or 7 further comprising:

creating from the mapping an RDF logical representation of the relational database data that conforms to the synthetic domain model.

26. The method of claim 4 further comprising materializing in the relational database the database contents as an RDF representation.

5. A method for the execution of digitally encoded SPARQL queries by a computer against the contents of a relational database with a corresponding relational schema stored in computer memory, said method comprising:

automatically generating a data structure representing a synthetic domain model that is putative ontology automatically created from the relational schema and a mapping from the relational schema to the synthetic domain model;

automatically translating the SPARQL query into a digitally encoded SQL query using the a subsent of the contents of the mapping data structure, where the arguments of the SQL query are the RDF representation of the relational database data contents; and

executing the translated SQL query on the relational database system.

6. The method of claim 5 further comprising:

formatting the SQL query to meet the requirements of the SPARQL query.

7. A method for executing a digitally encoded first query expressed in a semantic query language by a computer on data organized as a relational database with a corresponding schema stored in a computer system comprising:

automatically generating a data structure representing a mapping of the relational database schema to a synthetic domain model that is a putative ontology automatically created from the relational schema;

creating a VIEW construct taking as arguments values derived from a subset of the contents of the mapping data structure;

automatically translating the first query into a second digitally encoded relational query that contains an input parameter comprised of a reference to the VIEW construct.

8. The method of claim 7 where the VIEW construct is logical.

9. The method of claim 7 where the VIEW construct has been materialized.

27. A computer system comprised of at least one central processing unit operatively connected to at least one memory device for executing a digitally encoded database query received by the system that is comprised of a representation expressed in a first query language where the first query language operates on a first type of data organized as a graph, where the system is adapted to execute the received query on a second type of data organized as a relational database with a corresponding database schema, by using the computer system to automatically translate the query representation expressed in the first query language into a query representation expressed in a second query language that operates on the relational database by using a subset of the contents of a data structure at least part of which is stored in the at least one memory device, said contents of said data structure representing a mapping from the relational database schema to a synthetic domain model that is a putative ontology automatically created from the relational schema.

28. A computer system comprised of at least one central processing unit operatively connected to at least one memory device for executing a digitally encoded database query received by the system, said query being comprised of a representation expressed in a semantic query language on the contents of a pre-existing relational database with a corresponding relational database schema, said system comprising:

A module adapted to automatically generate a data structure at least partially stored in the at least one memory device representing a mapping of the pre-existing relational database schema from the relational database schema to a synthetic domain model that is a putative ontology automatically created from the relational schema; and

A module adapted to automatically translate the received query into a second digitally encoded relational query by using a subset of the contents of the mapping-data structure to determine component substitutions of at least one of the components comprising the received query in order to create the second relational query translation.

29. The system of claim 28 further comprising a module adapted to execute the second relational query on the pre-existing relational database and output a result.