System and method for data integration using multi-dimensional, associative unique identifiers

• US 7,634,482 B2
• Filed: 07/12/2004
• Issued: 12/15/2009
• Est. Priority Date: 07/11/2003
• Status: Active Grant

• Current Assignee: Ids Global, Inc.
• Original Assignee: Ids Global, Inc.
• Inventors: Mukherjee, Arka, Margolin, Oleg
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Khoshnoodi; Fariborz
• Application Number:
  US10/889,229
• Publication Number:
  US 20050097111A1
• Time in Patent Office: 1,982 Days
• Field of Search: 707/2, 707/3, 707/4, 707/6, 707/7, 707/100
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

A system and method for associating data objects utilizing unique identifiers is provided. Data objects are modeled utilizing a data object ontology. Unique identifiers for instances of each data object are calculated utilizing a selection of unique attributes of the data object ontology. Data objects from multiple data sources can be integrated utilizing the unique identifiers for each data object.

Claims

1. A computer implemented method including a plurality of data objects stored in a plurality databases, a method implemented in instructions executed by a computer processor for processing data stored on a database, the method comprising:

associating each a data object in the plurality of data objects in the plurality of databases with a data object ontology corresponding to a hierarchy data object of dimensions of the data object, wherein the hierarchy comprises having at least one source dimension; and

one or more target dimensions dependent on each source dimension and wherein each dimension associated with a specific attribute data; and

for each data object of the plurality of data objects;

based upon values of specific attribute data of a subset of data object dimensions associated with the data object comprising a source dimension and one or more dependent target dimensions;

providing the unique identifier to identify the data object;

calculating and associating a unique identifier for each data object based on applying a hashing algorithm executed by the computer processor to data stored on the database to a selected set of the objects'"'"' dimensions having the highest affinity metric values, and wherein the affinity metric for each dimension is calculated as a weighted sum of temporal invariance of a dimension and a uniqueness metric, wherein;

the temporal invariance of the object dimension is calculated as the minimum of the temporal invariance of any value appearing in the object dimension, the temporal invariance being calculated as the ratio of the number of times the value of the object dimension has changed over the total number of times the value of the dimension is observed in the databases over a period of time; and

the uniqueness metric capturing whether data values in an object dimension are unique.

2. The method as recited in claim 1, wherein the data object ontology includes an invariance strength identifier for each source and one or more dependent target dimensions in the hierarchy of data object dimensions.

3. The method as recited in claim 2, wherein calculating a unique identifier includes selecting a subset of data objects corresponding to the invariance strength identifier.

4. The method as recited in claim 2, wherein calculating a unique identifier includes selecting a subset of data objects corresponding to a uniqueness metric for each dimension in the hierarchy of data object dimensions.

5. The method as recited in claim 1, further comprising integrating data from a plurality of data sources utilizing the unique identifier.

6. The method as recited in claim 1, wherein integrating data includes generating a table mapping unique identifiers to identifications of data in the data sources.

7. The method as recited in claim 1, wherein the data objects correspond to business objects and wherein the data object ontology corresponds to a business object ontology.

8. The method as recited in claim 1, wherein calculating a unique identifier includes creating a hash of the selection of a subset of data object dimensions.

9. A computer implemented method including a plurality of data objects stored in a plurality of databases, a method implemented in instructions executed by a computer processor for processing data stored on a database, the method comprising:

associating each data object in the plurality of databases with a data object ontology corresponding to a hierarchy of data object dimensions of the data object, wherein the hierarchy comprising at least one source dimension and one or more target dimensions dependent on each source dimension and wherein each dimension associated with a specific attribute data and wherein the data object ontology includes a invariance strength identifier computed using a hashing function for a source and target dimension in the hierarchy of data object dimensions executed by the computer processor; and

for each of the plurality of data objects;

calculating a unique identifier for a data object based upon values of specific attribute data of a subset of data object dimensions associated with the data object, said dimensions comprising a source dimension and one or more dependent target dimensions having a high affinity metric; and

providing the unique identifier to identify the data object;

wherein the affinity metric for each dimension is calculated as a weighted sum of a temporal invariance of a dimension and a uniqueness metric, wherein;

the temporal invariance of the dimension is calculated as the minimum of the temporal invariance of any value appearing in the dimension, the temporal invariance being calculated as the ratio of the number of times the value of the dimension has changed over the total number of times the value of the dimension is observed in a period of time; and

the uniqueness metric capturing whether data values in the dimension are unique.

10. The method as recited in claim 9, further comprising integrating data from a plurality of data sources utilizing the unique identifier.

11. The method as recited in claim 9, wherein integrating data includes generating a table mapping unique identifiers to identifications of data in the data sources.

12. The method as recited in claim 9, wherein the data objects correspond to business objects and wherein the data object ontology corresponds to a business object ontology.

13. The method as recited in claim 9, wherein calculating a unique identifier includes creating a hash of the selection of a subset of data object dimensions.

14. The method as recited in claim 9, wherein the affinity metric corresponds to a combination of an invariance strength identifier and a uniqueness metric.

15. The method as recited in claim 9, wherein the affinity metric further comprises a weighted combination of the invariance strength identifier and the uniqueness metric.

16. A computing system comprised of modules stored on a database including a plurality of data objects modules for processing data objects stored in databases, the system comprising:

a processor;

a plurality of databases including data object dimension data corresponding to the plurality of data objects;

a data object ontology corresponding to a hierarchy of data object dimensions, wherein the hierarchy comprises at least one source dimension and one or more target dimensions dependent on each source dimension and wherein each dimension is associated with a specific attribute data of each data object; and

a data integration application implemented in instructions accessed from the database and executed by a computer processor configured to;

obtain the data object dimension data; and

, for each data object;

calculate a unique identifier for a data object based upon the temporal invariance and uniqueness of the values of specific attribute data of a subset of data object dimensions associated with the data object comprising a source dimension and one or more dependent target dimensions;

wherein the temporal invariance of an object dimension is calculated as the minimum of the temporal invariance of any value appearing in the object dimension, the temporal invariance being calculated as the ratio of the number of times the value of the dimension has changed over the total number of times the value of the dimension is observed in a period of time; and

integrate the data object dimension data according to the calculated unique identifier; and

provide the unique identifier to identify the data object.

17. The system as recited in claim 16, wherein the unique identifier is calculated from an invariance strength identifier for a source and target dimension in the hierarchy of data objects dimensions.

18. The system as recited in claim 17, wherein the unique identifier is calculated from a uniqueness metric for each dimension in the hierarchy of data objects dimensions.

19. The system as recited in claim 18, wherein the unique identifier is calculated from a weighted combination of the invariance strength identifier and the uniqueness metric.

20. The system as recited in claim 16, wherein the data integration application maintains a table mapping unique identifiers to identifications of data in the data sources.

21. The system as recited in claim 16, wherein the data objects correspond to business objects and wherein the data object ontology corresponds to a business object ontology.