Methods and systems for managing a data mining model

• US 9,043,267 B2
• Filed: 07/30/2007
• Issued: 05/26/2015
• Est. Priority Date: 07/30/2007
• Status: Expired due to Fees

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Kozlov, Alexander, Li, Jun, Ari, Ismail
• Primary Examiner(s): Fan, Shiow-Jy
• Application Number:
  US11/830,791
• Publication Number:
  US 20090037363A1
• Time in Patent Office: 2,857 Days
• Field of Search: 707/1, 707/E17.005
• US Class Current: 707/1
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

Methods, systems, and a computer readable medium storing a computer executable program for managing a data mining model are disclosed. A first notification associated with a data mining model is received at a business process system. A second notification associated with the data mining model is transmitted from the business process system responsive to the first notification. A determination is made regarding whether a first response operable to define an association between the data mining model and business data has been received at the business process system responsive to the second notification. A command is issued from the business process system to update model metadata associated with the data mining model based on the determination.

Claims

1. A method of managing a data mining model, the method comprising:

receiving a first notification associated with a data mining model at a business process system, the first notification received from a model developer system;

transmitting, in response to the first notification, a second notification associated with the data mining model from the business process system to a business analyst system;

determining whether a first response operable to define an association between the data mining model and business data has been received from the business analyst system at the business process system, the first response responsive to the second notification;

issuing a command from the business process system to update model specific metadata associated with the data mining model based on the determination;

issuing from a model evaluation system a command to evaluate the data mining model, the model evaluation system generating a model performance status associated with the data mining model; and

issuing a command from the business process system to update model specific metadata associated with the data mining model based on the model performance status.

2. The method of claim 1, wherein receiving a first notification associated with a data mining model comprises receiving a first notification selected from a group consisting of a model created notification and a model to be expired notification.

3. The method of claim 1, wherein transmitting a second notification associated with the data mining model comprises transmitting a second notification selected from a group consisting of a model created notification and a model to be expired notification.

4. The method of claim 1, further comprising:

retrieving at least one model specific keyword associated with the data mining model;

identifying a business analyst associated with the at least one of the model specific keywords; and

transmitting the second notification associated with the data mining model from the business process system to the business analyst system associated with the identified business analyst.

5. The method of claim 1, wherein determining whether a first response operable to define an association between the data mining model and business data has been received at the business process system comprises determining whether a first response operable to define an association selected from a group consisting of a model referenced association and a model unreferenced association has been received at the business process system.

6. The method of claim 1, further comprising:

determining whether the model performance status is a model deteriorated status; and

issuing a model performance notification associated with the data mining model from the business process system to the model developer system based on the determination.

7. The method of claim 6, further comprising:

receiving a third notification associated with the data mining model from the model developer system at the business process system;

making a first determination regarding whether the third notification is a model updated notification associated with the data mining model;

updating the model specific metadata associated with the data mining model based on the first determination;

making a second determination regarding whether the third notification includes a model expiration status associated with the data mining model; and

issuing a model expiration status notification from the business process system to the business analyst system based on the second determination.

8. The method of claim 1, further comprising providing a model metadata repository operable to be communicatively coupled to the business process system and operable to store model specific metadata associated with the data mining model.

9. The method of claim 8, wherein providing a model metadata repository comprises providing a model metadata repository operable to store model specific metadata selected from a group consisting of a basic model information, a model schema definition, a model assumption, a model specific keyword, a model performance evaluation definition, a model performance event trigger, a model business rule dependency, and an inter-model dependency.

10. A non-transitory computer readable medium for storing a computer executable program for managing a data mining model, the computer readable medium comprising:

computer readable code for receiving a first notification associated with a data mining model at a business process system, the first notification received from a model developer system;

computer readable code for transmitting, in response to the first notification, a second notification associated with the data mining model from the business process system to a business analyst system;

computer readable code for determining whether a first response operable to define an association between the data mining model and business data has been received from the business analyst system at the business process system, the first response responsive to the second notification;

computer readable code for issuing a command from the business process system to update model specific metadata associated with the data mining model based on the determination;

computer readable code for issuing from a model evaluation system a command to evaluate the data mining model, the model evaluation system generating a model performance status associated with the data mining model; and

computer readable code for issuing a command from the business process system to update model specific metadata associated with the data mining model based on the model performance status.

11. The computer readable medium of claim 10, wherein the computer readable code for receiving a first notification associated with a data mining model comprises computer readable code for receiving a first notification selected from a group consisting of a model created notification, and a model to be expired notification.

12. The computer readable medium of claim 10, wherein the computer readable code for transmitting a second notification associated with the data mining model comprises computer readable code for transmitting a second notification selected from a group consisting of a model created notification and a model to be expired notification.

13. The computer readable medium of claim 10, further comprising:

computer readable code for retrieving at least one model specific keyword associated with the data mining model;

computer readable code for identifying a business analyst associated with the at least one of the model specific keywords; and

computer readable code for transmitting the second notification associated with the data mining model from the business process system to the business analyst system associated with the identified business analyst.

14. The computer readable medium of claim 10, wherein the computer readable code for determining whether a first response operable to define an association between the data mining model and business data has been received at the business process system comprises computer readable code for determining whether a first response operable to define an association selected from a group consisting of a model referenced association and a model unreferenced association has been received at the business process system.

15. The computer readable medium of claim 10, further comprising:

computer readable code for determining whether the model performance status is a model deteriorated status; and

computer readable code for issuing a model performance notification associated with the data mining model from the business process system to the model developer system based on the determination.

16. The computer readable medium of claim 15, further comprising:

computer readable code for receiving a third notification associated with the data mining model from the model developer system at the business process system;

computer readable code for making a first determination regarding whether the third notification is a model updated notification associated with the data mining model;

computer readable code for updating the model specific metadata associated with the data mining model based on the first determination;

computer readable code for making a second determination regarding whether the third notification includes a model expiration status associated with the data mining model; and

computer readable code for issuing a model expiration status notification from the business process system to the business analyst system based on the second determination.

17. The computer readable medium of claim 10, further comprising computer readable code for storing model specific metadata associated with the data mining model at a model metadata repository, the model metadata repository being communicatively coupled to the business process system.

18. The computer readable medium of claim 17, wherein the computer readable code for storing model specific metadata associated with the data mining model at a model metadata repository comprises storing model specific metadata selected from a group consisting of a basic model information, a model schema definition, a model assumption, a model specific keyword, a model performance evaluation definition, a model performance event trigger, a model business rule dependency, and an inter-model dependency.

19. A system for managing a data mining model, comprising:

a processor; and

a memory, comprising computer-executable instructions that, when executed by the processor, cause the processor to;

receive a first notification associated with a data mining model at a business process system, the first notification received from a model developer system;

transmit, in response to the first notification, a second notification associated with the data mining model from the business process system to a business analyst system;

determine whether a first response operable to define an association between the data mining model and business data has been received from the business analyst system at the business process system, the first response responsive to the second notification;

issue a command from the business process system to update model metadata associated with the data mining model based on the determination;

issue from a model evaluation system a command to evaluate the data mining model, the model evaluation system generating a model performance status associated with the data mining model; and

issue a command from the business process system to update model specific metadata associated with the data mining model based on the model performance status.