Infrastructure and architecture for development and execution of predictive models

• US 7,953,762 B2
• Filed: 04/09/2008
• Issued: 05/31/2011
• Est. Priority Date: 04/09/2008
• Status: Active Grant

• Current Assignee: American Express Travel Related Services Company, Inc. (American Express Company)
• Original Assignee: American Express Travel Related Services Company, Inc. (American Express Company)
• Inventors: Sacheti, Sandeep, Murthy, Narasimha, Thomas, Deep, Durvasula, Sastry VSM, Von Wolff, Karl, Agrawal, Sanjay S.
• Primary Examiner(s): Breene; John
• Assistant Examiner(s): Ly; Anh
• Application Number:
  US12/100,225
• Publication Number:
  US 20090259685A1
• Time in Patent Office: 1,147 Days
• Field of Search: 707/736, 707/810, 707/713, 707/802, 707/805, 702/19, 702/20, 703/11, 703/22, 705/7, 705/14.52, 705/35
• US Class Current: 707/802
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: Y10S 707/948 Product or catalog

Abstract

A system that enables development and execution of predictive models comprises a centralized data management system, a data extraction tool a model validation tool and a model execution tool. In embodiments, a data management system includes a data management server that can be accessed via a web browser that stores data. An extraction tool includes a data filter adapted to filter data based on, for example, a population criteria, a sample size, and a date range criteria. A model validation tool validates the model. A model execution tool allows a user to score the model.

Claims

1. A method comprising:

displaying, by a computer-based system for generating a plurality of predictive models and on a user interface, first data associated with a first group of consumers,

manipulating, by the computer-based system, the first data by at least one of;

filtering the first data based on a sample size, filtering the first data based on an uploaded list, filtering the first data using mathematical and logical operations, and filtering the first data based on a credit limit;

receiving, by the computer-based system and through the user interface, second data associated with a second group of consumers and at least one of;

a score destination and an update interval;

simultaneously generating and executing, by the computer-based system, computer logic that represents the plurality of predictive models, wherein at least one of the predictive models in the plurality of predictive models is based upon the analysis and manipulation of the first data, and wherein each model is given a score representing the accuracy of the model;

transmitting, by the computer-based system, to each consumer in the second group of consumers a score that indicates the likelihood that each consumer in the group will respond to an offer; and

tracking, by the computer-based system, a behavior and performance of the plurality of predictive models using a Population Stability Index (“

PSI”

).

2. The method of claim 1, further comprising simultaneously generating and executing a plurality of predictive models.

3. The method of claim 1, further comprising scoring during specified intervals the predictive model, wherein the score represents the accuracy of the model.

4. The method of claim 1, further comprising storing, by the computer-based system, Structured Query Language (“

SQL”

) queries in a first database.

5. The method of claim 4, further comprising storing, by the computer-based system, Statistical Data Analysis (“

SAS”

) code in a second database.

6. The method of claim 5, wherein the first data is queried using the SAS code.

7. The method of claim 5, wherein the first data is queried using the SQL queries.

8. The method of claim 1, further comprising displaying, by the computer-based system and using a searchable data dictionary, policy usage criteria associated with the first data.

9. The method of claim 1, further comprising checking, by the computer-based system and using an intelligent code parser, the computer logic for errors.

10. The method of claim 1, further comprising collecting and standardizing the first data.

11. The method of claim 1, further comprising optimizing, by the computer-based system, how the first data are accepted by way of a self-learning process.

12. The method of claim 1, further comprising validating, by the computer-based system, the predictive model prior to deployment of the predictive model.

13. The method of claim 12, further comprising presenting, through the user interface, a standardized list of rules associated with a data testing procedure.

14. The method of claim 1, further comprising updating, by the computer-based system and through the user interface, the predictive model after the predictive model has been deployed.

15. The method of claim 1, further comprising maintaining, by the computer-based system, a library that includes a plurality of the predictive models.

16. The method of claim 1, further comprising grouping, by the computer based system, a plurality of consumers based on a spending attribute associated with the plurality of consumers to create the first group of consumers.

17. An article of manufacture including a non-transitory, tangible computer-readable storage medium having instructions stored thereon for generating a plurality of predictive models that, in response to execution by a computer-based system for generating the plurality of predictive models, cause the computer-based system to perform operations comprising:

displaying, by the computer-based system and on a user interface, first data associated with a first group of consumers,

manipulating, by the computer-based system, the first data by at least one of;

filtering the first data based on a sample size, filtering the first data based on an uploaded list, filtering the first data using mathematical and logical operations, and filtering the first data based on a credit limit;

receiving, by the computer-based system and through the user interface, second data associated with a second group of consumers and at least one of;

a score destination and an update interval;

simultaneously generating and executing, by the computer-based system, computer logic that represents the plurality of predictive models, wherein at least one of the predictive models in the plurality of predictive models is based upon the analysis and manipulation of the first data, and wherein each model is given a score representing the accuracy of the model;

transmitting, by the computer-based system, to each consumer in the second group of consumers a score that indicates the likelihood that each consumer in the group will respond to an offer; and

tracking, by the computer-based system, a behavior and performance of the plurality of predictive models using a Population Stability Index (“

PSI”

).

18. A system comprising:

a tangible, non-transitory memory communicating with a processor for generating a plurality of predictive models,

the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations comprising;

displaying, by the processor and on a user interface, first data associated with a first group of consumers,

manipulating, by the processor, the first data by at least one of;

filtering the first data based on a sample size, filtering the first data based on an uploaded list, filtering the first data using mathematical and logical operations, and filtering the first data based on a credit limit;

receiving, by the processor and through the user interface, second data associated with a second group of consumers and at least one of;

a score destination and an update interval;

simultaneously generating and executing, by the processor, computer logic that represents the plurality of predictive models, wherein at least one of the predictive models in the plurality of predictive models is based upon the analysis and manipulation of the first data, and wherein each model is given a score representing the accuracy of the model;

transmitting, by the processor, to each consumer in the second group of consumers a score that indicates the likelihood that each consumer in the group will respond to an offer; and

tracking, by the processor, a behavior and performance of the plurality of predictive models using a Population Stability Index (“

PSI”

).