Data mining framework using a signature associated with an algorithm

• US 7,024,417 B1
• Filed: 11/14/2002
• Issued: 04/04/2006
• Est. Priority Date: 11/14/2002
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Hyperion Solutions Corporation (Oracle Corporation)
• Inventors: Rodny, Uri, Russakovsky, Alexander
• Primary Examiner(s): Pham, Khanh
• Application Number:
  US10/295,593
• Time in Patent Office: 1,237 Days
• Field of Search: 707/6, 707/101, 707/103.R, 717104-109
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/258 Data format conversion from or to a database
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure

Abstract

A framework is provided that enables data mining algorithms to be plugged into it without any change to algorithm software implementations, while still providing all the standard data mining tasks. It may be implemented by the data source provider. It also then allows for the complete separation of data storage and algorithms. When the user initiates a mining session and picks an algorithm for build task or a model for an apply or test task, the framework may become responsible for preparing a set of “prompts” to the user asking him to provide some expression which is specific to the particular kind of data the user is working with.

Claims

1. A method for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the method comprising:

retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm; and

creating a template for said the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

executing said template to create a mapping between said one or more coefficients and said one or more model values.

2. The method of claim 1, further comprising:

verifying said signature.

3. The method of claim 2, wherein said verifying comprises:

comparing said signature to a predefined standard format for said signature; and

rejecting said signature if said signature does not match said predefined standard format.

4. The method of claim 3, wherein said predefined standard format is a document type definition.

5. The method of claim 1, further comprising:

generating a graphical dialog based on said template, said graphical dialog allowing a user to initialize required parameters.

6. The method of claim 5, further comprising:

dynamically creating accessor and domain objects in response to user-initialized required parameters; and

assembling said accessor and domain objects into a data provider object.

7. The method of claim 6, further comprising calling the algorithm using said data provider object as an argument.

8. The method of claim 6, wherein said graphical dialog prompts a user to provide a mapping from terms of the algorithm to an actual data source.

9. The method of claim 8, wherein said terms of the algorithm include accessors and domains.

10. The method of claim 9, wherein said graphical dialog further prompts a user to provide a source specific linguistic expression for each axis in each accessor.

11. The method of claim 10, wherein said source specific linguistic expression is a source specific query object.

12. A method for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the method comprising:

retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm;

creating a template for the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

executing said template to create a mapping between said one or more coefficients and said one or more model values, said execution generating a set of prompts asking said user to provide some expression specific to a data source said user is working with.

13. An apparatus for data mining using an algorithm, the algorithm having, a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the apparatus comprising:

a signature information field receiver configured to retrieve a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm;

a task template creator coupled to said signature information field receiver, wherein the task template creator is configured to create a template for the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

an executing module coupled to the task template creator, wherein the execution module is configured to execute said template to create a mapping between said one or more coefficients and said one or more model values.

14. The apparatus of claim 13, further comprising a signature verifier coupled to said signature information field receiver.

15. The apparatus of claim 13, further comprising a graphical dialog generator coupled to said task template creator.

16. The apparatus of claim 15, further comprising:

an accessor and domain objects dynamic creator coupled to said graphic dialog generator; and

an accessor and domain object data provider object assembler coupled to said accessor and domain objects dynamic creator.

17. The apparatus of claim 16, further comprising an algorithm caller coupled to said accessor and domain object data provider object assembler.

18. An apparatus for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the apparatus comprising:

means for retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm;

means for creating a template for the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

means for executing said template to create a mapping between said one or more coefficients and said one or more model values.

19. The apparatus of claim 18, further comprising:

means for verifying said signature.

20. The apparatus of claim 19, wherein said verifying comprises:

means for comparing said signature to a predefined standard format for said signature; and

means for rejecting said signature if said signature does not match said predefined standard format.

21. The apparatus of claim 20, wherein said predefined standard format is a document type definition.

22. The apparatus of claim 18, further comprising:

means for generating a graphical dialog based on said template, said graphical dialog allowing a user to initialize required parameters.

23. The apparatus of claim 22, further comprising:

means for dynamically creating accessor and domain objects in response to user-initialized required parameters; and

means for assembling said accessor and domain objects into a data provider object.

24. The apparatus of claim 23, further comprising means for calling the algorithm using said data provider object as an argument.

25. The apparatus of claim 24, wherein said graphical dialog prompts a user to provide a mapping from terms of the algorithm to an actual data source.

26. The apparatus of claim 25, wherein said terms of the algorithm include accessors and domains.

27. The apparatus of claim 26, wherein said graphical dialog further prompts a user to provide a source specific linguistic expression for each axis in each accessor.

28. The apparatus of claim 27, wherein said source specific linguistic expression is a source specific query object.

29. An apparatus for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the apparatus comprising:

means for retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm;

means for creating a template for the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

means for executing said template to create a mapping between said one or more coefficients and said one or more model values, said execution generating a set of prompts asking said user to provide some expression specific to a data source said user is working with.

30. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform a method for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the method comprising:

retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm; and

creating a template for the build task based on said signature, said template indicating one or more of said parameters that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

executing said template to create a mapping between said one or more coefficients and said one or more model values.

31. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform a method for data mining using an algorithm, the algorithm having a build task, a test task, and an apply task, each task having a number of parameters, each parameter having a type, the method comprising:

retrieving a signature associated with the algorithm, said signature including, for the build task, the number of parameters and the type of each parameter associated with said task, as well as an information field for each parameter associated with said task, said information field indicating the meaning and/or recommended usage of said parameter, said signature also including, for the build task, one or more coefficients for the algorithm;

creating a template for the build task based on said signature, said template indicating one or more of said parameters fields that need to be initialized by a user to invoke said task and one or more model values that are to be derived from a data set; and

executing said template to create a mapping between said one or more coefficients and said one or more model values, said execution generating a set of prompts asking said user to provide some expression specific to a data source said user is working with.