System for ontology-based semantic matching in a relational database system

• US 7,328,209 B2
• Filed: 08/11/2004
• Issued: 02/05/2008
• Est. Priority Date: 08/11/2004
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Chong, Eugene Inseok, Das, Souripriya, Eadon, George, Srinivasan, Jagannathan
• Primary Examiner(s): Gaffin; Jeffrey
• Assistant Examiner(s): PULLIAM, CHRISTYANN R
• Application Number:
  US10/916,547
• Publication Number:
  US 20060036592A1
• Time in Patent Office: 1,273 Days
• Field of Search: 707/3, 707/4
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2445 Data retrieval commands; View definitions
• CPC Class Codes: G06F 16/2448 for particular applications; for extensibility, e.g. user defined types
• CPC Class Codes: G06F 16/24566 Recursive queries
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

The method for processing data in a relational database wherein ontology data that specifies terms and relationships between pairs of said terms expressed in an OWL document is stored in the database, database queries that include a semantic matching operator are formed which identify the ontology data and further specify a stated relationship between two input terms, and the query is executed to invoke the semantic matching operator to determine if the two input terms are related by the stated relationship by consulting said ontology data.

Claims

1. A method of processing data stored in a relational database management system comprising, in combination, the steps of:

storing ontology data to which the relational database management system has access that specifies terms and relationships between pairs of said terms,

forming a database query in a query language that is native to the relational database management system, the database query including a semantic matching operator defined in the native query language that identifies said ontology data and a further specifies a stated relationship between two input terms, and

executing said query in the relational database management system to invoke said semantic matching operator to consult said ontology data and return a result to said executing query which indicates whether said two input terms are related by the stated relationship.

2. The method of processing data stored in a relational database management system as set forth in claim 1 wherein said ontology data specifies relationships having a plurality of types between said pairs of terms and wherein said semantic matching operator further specifies a type of the plurality of types for the stated relationship.

3. The method of processing data stored in a relational database management system as set forth in claim 2 wherein said ontology data is identified by an ontology name value and wherein said semantic matching operator specifies said ontology data using said ontology name value.

4. A method of identifying target values stored in one or more selected columns of selected tables in a relational database management system where said target values have relationships to ontology values that are defined in ontology data to which the relational database management system has access, said method comprising:

forming and executing an SQL query statement in the relational database management system that includes an identification of a semantic matching operator defined in the relational database system that is invoked to identify said target values by consulting said ontology data and returning a result to said executing SQL query statement which indicates whether said target values are related to said ontology values.

5. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 4 wherein said identification of a semantic matching operator is a user-defined operator defined in said relational database management system that extends the pre-existing SQL syntax employed by said relational database management system.

6. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 4 wherein said SQL query statement further includes the identification of said one or more selected columns.

7. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 6 wherein said semantic matching operator is invoked to match target values in said one or more selected columns one with said ontology values by consulting said ontology data.

8. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 4 wherein said ontology data specifies relationships having a plurality of types between said target values and said ontology values and wherein said semantic matching operator further specifies a type of the plurality of types for the relationship.

9. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 8 wherein at least some of said relationships of different types are transitive and wherein said semantic matching operator is invoked to search transitive pathways connected by said transitive relationships.

10. The method of identifying target values stored in one or more selected columns of selected tables in a relational database management system as set forth in claim 9 wherein said semantic matching operator further computes a measure of how closely said target values are related to said ontology values by measuring said transitive pathways.

11. A method of performing an ontology-based matching operation during the execution of a query in a relational database management system, said query being defined by a native structured query language statement, and said method comprising, in combination, the steps of:

creating and storing in said relational database management system a semantic matching operator that determines whether two input terms are related by a specified relationship as defined by specified ontology data that is accessible to said a relational database management system, and

constructing and executing a query expressed in the native structured query language in a relational database management system, the query including an identification of said semantic matching operator.

12. The method of performing an ontology-based matching operation as set forth in claim 11 wherein said specified ontology data specifies relationships having types belonging to a plurality of types between said input terms and wherein said identification of said semantic matching operator further specifies a type of the plurality for the relationship.

13. The method of performing an ontology-based matching operation as set forth in claim 12 wherein at least some of said plurality of types of relationships are properties each property specifying a relationship between one term and one or more other terms and wherein said semantic matching operator specifies a property of said properties.

14. The method of performing an ontology-based matching operation as set forth in claim 13 wherein said specified ontology data further specifies that at least one of said properties is a subproperty of another of said properties and wherein said semantic matching operator specifies properties that are subproperties of an identified property.

15. The method of performing an ontology-based matching operation as set forth in claim 13 wherein said properties define transitive relationships and wherein said semantic matching operator infers new relationships based on said transitive relationships during execution.