Method and system for the visual presentation of data mining models

• US 7,512,623 B2
• Filed: 12/23/2003
• Issued: 03/31/2009
• Est. Priority Date: 07/06/2001
• Status: Active Grant

• Current Assignee: Angoss Software Corporation (Altair Engineering Incorporated)
• Original Assignee: Angoss Software Corporation (Altair Engineering Incorporated)
• Inventors: Ono, Ken, Apps, Eric
• Primary Examiner(s): Trujillo; James
• Assistant Examiner(s): Casanova; Jorge A
• Application Number:
  US10/743,324
• Publication Number:
  US 20040215599A1
• Time in Patent Office: 1,925 Days
• Field of Search: 707/3, 707/102
• US Class Current: 1/1
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method for delivering presentations associated with data mining models. The method includes the steps of: a) storing the data mining models and the presentations in a repository in a data mining system; b) associating ones of the data mining models with ones of the presentations through references within ones of the presentations; c) receiving a request from a user for a presentation; d) processing the request to determine an appropriate presentation corresponding to the request; the request including a customer identification; the request including an event identification; and, e) delivering the appropriate presentation to the user.

Claims

1. A data mining system for delivering presentations associated with data mining models, said data mining system comprising:

a repository to store said data mining models, customer attributes, and presentation definitions;

means to edit said data mining models, said presentation definitions, and said customer attributes;

means to generate a presentation to deliver to a customer system;

wherein said means to generate includes an analytic decision engine system including model presentation services and scoring services modules; and

,

means to receive inputs from said customer system and to deliver said presentation to said customer system;

wherein said inputs include a customer identification and a presentation definition identification;

wherein said means to generate selects a presentation definition using said presentation definition identification and selects a customer attribute using said customer identification;

wherein said presentation definition includes a reference to a data mining model and one or more rules;

and,

wherein said means to generate applies said data mining model and said one or more rules to said customer attribute to produce an outcome for display in said presentation according to a format included in said presentation definition.

2. The system of claim 1 wherein said data mining models include one or more of simple scores, complex scores, static scores, dynamic scores, rules, rules sets, and rules systems.

3. The system of claim 1 wherein said presentation definitions include defined sets of content including one or more of references to said data mining models, scores, rules, said customer attributes, activity suggestions, to do lists, pop-up windows, HTML segments, and XML scripts.

4. The system of claim 1 wherein said presentations include one or more of simple presentations, range based presentations, and complex presentations.

5. The system of claim 1 wherein said repository is a database having a database schema and a database management system.

6. The system of claim 1 wherein said means to edit said data mining models, said presentation definitions, and said customer attributes includes a data mining console system.

7. The system of claim 1 wherein said means to edit said data mining models, said presentation definitions, and said customer attributes includes one or more of text editors, scripting tools, web development tools, and HTML editors.

8. The system of claim 1 wherein said means to receive inputs from said customer system and to deliver said presentations to said customer system includes a network.

9. The system of claim 1 wherein said customer system includes one or more data mining management consoles integrated with or connected over a network.

10. The system of claim 1 wherein said customer system includes one or more applications running on a computer, system, or other appliance integrated with or connected over a network.

11. The system of claim 1 wherein said customer system includes one or more thin or thick clients in a client-server or browser-server environment integrated with or connected over a network.

12. The system of claim 1 wherein said inputs include one or more of said customer attributes.

13. The system of claim 12 wherein said data mining system includes one or more reporting systems.

14. The system of claim 13 wherein said reporting systems include one or more of outcome monitoring systems and presentation usage monitoring systems.

15. The system of claims 14 wherein said reporting systems provide one or more of said inputs.

16. The system of claim 1 wherein said data mining system includes one or more of servers, memory devices, processing units, input devices, output devices, and display devices.

17. The system of claim 8 wherein said network is an Internet network.

18. A method for generating presentations of outcomes associated with data mining models for display on a display screen, comprising:

storing one or more presentation definitions in a repository;

wherein the presentation definitions contain one or more rules, one or more references to data mining models, and one or more references to customer attributes;

wherein the presentation definitions have respective formatting information for formatting presentations of outcomes;

wherein the customer attributes are name-value pairs;

wherein the data mining models generate scores based on one or more of the customer attributes; and

, wherein the rules are for branching on one or more of the scores and the customer attributes to choose between outcomes;

receiving a request for a presentation for an outcome, the request including a presentation definition identification and one or more of a customer identification and one or more customer attributes;

wherein the presentation definition identification is a key to a presentation definition stored in the repository; and

, wherein the customer identification is a key to one or more customer attributes stored in the repository;

in response to the request, retrieving a presentation definition from among the one or more presentation definitions using the presentation identification and retrieving one or more of the customer attributes from the repository using the customer identification;

generating one or more scores from the data mining models referenced by the presentation definition using one or more of the customer attributes provided by the request or retrieved from the repository;

determining the outcome by applying one or more rules in the presentation definition to one or more of the scores and the customer attributes; and

,

formatting the presentation for the outcome for display on the display screen using the formatting information contained in the presentation definition.

19. The method of claim 18 wherein the data mining models are one or more of a logistic regression, a decision tree, a neural network, a Bayesian network, a linear regression, a cluster model, a K-Means cluster model, an expectation maximizing cluster model, and an association rule.

20. The method of claim 18 wherein the customer attributes are one or more of age information and sex information.

21. The method of claim 18 wherein the formatting information is markup language formatting information.

22. The method of claim 18 and further comprising importing the data mining models and presentation definitions into the repository.

23. The method of claim 18 and further comprising receiving updated data mining models and updated presentation definitions to replace the data mining models and presentation definitions in the repository.

24. The method of claim 18 and further comprising receiving the one or more rules for the presentation definition.

25. The method of claim 18 wherein the scores are one or more of simple scores, complex scores, static scores, and dynamic scores.

26. The method of claim 18 wherein the outcomes provide one or more of defined sets of content, data mining model scores, activity suggestions, to-do lists, pop-up windows, HTML segments, extensible markup language (XML) scripts, and sets of computer instructions.

27. The method of claim 18 wherein the outcomes are one or more of simple messages, range based messages, and complex messages.

28. The method of claim 18 wherein the request is received over a network.

29. The method of claim 28 wherein the network is an Internet network.