Systems and methods for object-based modeling using hierarchical model objects

• US 8,417,739 B2
• Filed: 05/29/2009
• Issued: 04/09/2013
• Est. Priority Date: 05/29/2009
• Status: Active Grant

• Current Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Original Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Inventors: Williamson, Eric
• Primary Examiner(s): Ng, Amy
• Assistant Examiner(s): Khong, Alexander
• Application Number:
  US12/475,458
• Publication Number:
  US 20100306272A1
• Time in Patent Office: 1,411 Days
• Field of Search: 707/804, 707/810
• US Class Current: 707/804
• CPC Class Codes: G06F 16/282 Hierarchical databases, e.g. IMS, LDAP data stores or Lotus Notes

Abstract

Embodiments relate to systems and methods for object-based modeling using hierarchical model objects. A modeling client can host modeling logic and an application programming interface (API) to create, access, manipulate, and import/export modeling objects used in modeling applications, such as engineering, medical, financial, and other modeling platforms. In aspects, the source data accepted into the modeling client can include consumer or business-level applications, whose spreadsheet, database or other content can be extracted and encapsulated in extensible markup language (XML) or other object-oriented format. Various model objects can be extracted from source data, and aggregated into a hierarchical model object encapsulating and preserving the underlying object layers. In aspects, one or more of the model object layers can functionally combine at least two other layers, to relate levels of data in hierarchical fashion. The modeling client can exchange composite or other modeling objects with external platforms on a programmatic basis.

Claims

1. A method comprising:

executing an instance of a multi-dimensional relational database application, the multi-dimensional relational database application comprising a set of source data;

selecting a first portion of source data from the instance of the multi-dimensional relational data application to extract a first set of data elements and a first set of functional relationships describing relationships between values of the first set of data elements;

encapsulating the first set of data elements and the first set of functional relationships, in a first predictive model object layer;

selecting a second portion of source data to extract a second set of data elements and a second set of functional relationships corresponding to the second set of data elements;

encapsulating the second set of data elements and the second set of functional relationships, in a second predictive model object layer, wherein each of the first predictive model object layer and the second predictive model object layer reveal data elements of the other predictive model object layer;

aggregating, by a processor, the first predictive model object layer and the second predictive model object layer to create a hierarchical predictive model object, wherein a value predicted by the hierarchical predictive model object is derived from at least one of the data elements of at least one of the first predictive model object layer or the second predictive model object layer; and

generating a predicted output using data elements and functional relationships for each layer in the hierarchical predictive model object.

2. The method of claim 1, wherein the first predictive model object layer and the second predictive model object layer are selectively accessible via the hierarchical predictive model object.

3. The method of claim 1, wherein the first predictive model object layer and the second predictive model object layer comprise at least one component model object encoded in extensible markup language (XML) format.

4. The method of claim 1, wherein at least one of the first predictive model object layer or the second predictive model object layer comprises at least one data element from at least two other predictive model object layers of a plurality of predictive model object layers.

5. The method of claim 1, wherein a dimension in the multi-dimensional relational database comprises at least one of a row, a column, or a cell.

6. The method of claim 1, wherein the set of source data comprises at least one attribute, the at least one attribute comprising at least one of a function call parameter or a linkage parameter.

7. The method of claim 1, wherein the first predictive model object layer and the second predictive model object layer are independently updatable via at least one modeling module.

8. The method of claim 1, wherein the hierarchical predictive model object is exportable via an application programming interface (API).

9. The method of claim 1, wherein the set of source data is hosted in at least one of local storage of a client computer or remote storage.

10. A client system, comprising:

an interface to a set of source data; and

a processor coupled to the interface, the processor being configured to;

execute an instance of a multi-dimensional relational database application, the multi-dimensional relational database application comprising the set of source data;

select a first portion of source data from the instance of the multi-dimensional relational data application to extract a first set of data elements and a first set of functional relationships describing relationships between values of the first set of data elements;

encapsulate the first set of data elements and the first set of functional relationships, in a first predictive model object layer;

select a second portion of source data to extract a second set of data elements and a second set of functional relationships corresponding to the second set of data elements;

encapsulate the second set of data elements and the second set of functional relationships, in a second predictive model object layer, wherein each of the first predictive model object layer and the second predictive model object layer reveal data elements of the other predictive model object layer;

aggregate the first predictive model object layer and the second predictive model object layer to create a hierarchical predictive model object, wherein a value predicted by the hierarchical predictive model object is derived from at least one of the data elements of at least one of the first predictive model object layer or the second predictive model object layer; and

generate a predicted output using data elements and functional relationships for each layer in the hierarchical predictive model object.

11. The system of claim 10, wherein the first predictive model object layer and the second predictive model object layer are selectively accessible via the hierarchical predictive model object.

12. The system of claim 10, wherein the first predictive model object layer and the second predictive model object layer comprise at least one model object encoded in extensible markup language (XML) format.

13. The system of claim 10, wherein at least one of the first predictive model object layer or the second predictive model object layer comprises at least one data element from at least two other predictive model object layers of a plurality of predictive model object layers.

14. The system of claim 10, wherein a dimension in the multi-dimensional relational database comprises at least one of a row, a column, or a cell.

15. The system of claim 10, wherein the set of source data comprises at least one attribute, the at least one attribute comprising at least one of a function call parameter or a linkage parameter.

16. The system of claim 10, wherein the first predictive model object layer and the second predictive model object layer are independently updatable via at least one modeling module.

17. The system of claim 10, wherein the hierarchical predictive model object is exportable via an application programming interface (API).

18. The system of claim 10, where the set of source data is hosted in at least one of local storage of the client system or remote storage.

19. A non-transitory computer readable storage media including instructions, that when executed by a processor, cause the processor to perform a method comprising:

executing an instance of a multi-dimensional relational database application, the multi-dimensional relational database application comprising a set of source data;

selecting a first portion of source data from the instance of the multi-dimensional relational data application to extract a first set of data elements and a first set of functional relationships describing relationships between values of the first set of data elements;

encapsulating the first set of data elements and the first set of functional relationships, in a first predictive model object layer;

selecting a second portion of source data to extract a second set of data elements and a second set of functional relationships corresponding to the second set of data elements;

encapsulating the second set of data elements and the second set of functional relationships, in a second predictive model object layer, wherein each of the first predictive model object layer and the second predictive model object layer reveal data elements of the other predictive model object layer;

aggregating, by the processor, the first predictive model object layer and the second predictive model object layer to create a hierarchical predictive model object, wherein a value predicted by the hierarchical predictive model object is derived from at least one of the data elements of at least one of the first predictive model object layer or the second predictive model object layer; and

generating a predicted output using data elements and functional relationships for each layer in the hierarchical predictive model object.

20. The non-transitory computer readable storage media of claim 19, wherein the hierarchical predictive model object is exportable via an application programming interface (API).