Infrastructure and architecture for development and execution of predictive models

• US 9,684,869 B2
• Filed: 10/09/2015
• Issued: 06/20/2017
• 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: Thomas, Deep, Murthy, Narasimha, Sacheti, Sandeep, Oralevich, Mickey S.
• Primary Examiner(s): Fleurantin, Jean B
• Application Number:
  US14/879,343
• Publication Number:
  US 20160034819A1
• Time in Patent Office: 620 Days
• Field of Search: 707758, 707802, 707804
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/148 File search processing
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• 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

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 in the form of a flat file. An extraction tool extracts data. A model validation tool validates a model by scoring an analytical environment data set and a production environment data set. A model execution tool allows a user to select when and how often a model is scored.

Claims

1. A system comprising:

a processor;

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

converting, by the processor using an extraction tool having a processing Structured Query Language (SQL) macro, Structured Query Language (SQL) variables to Statistical Analysis System (SAS) variables with standardized and consistent Statistical Analysis System (SAS) variable names in SAS flat files on a data management server;

extracting, by the processor, data from the SAS flat files on the data management server based at least in part on contents of customer files respectively associated with customers, wherein the data comprises customer variables,

analyzing, by the processor, the customer variables in the customer files to determine trends in customer behavior;

determining, by the processor, the customer variables to use in a plurality of models, wherein the customer variables referenced in a variable file are determined by the plurality of models that use a particular variable;

determining, by the processor, model logic based on the customer variables and the trends in customer behavior;

enabling, by the processor, creation of the plurality of models based on the model logic; and

delivering, by the processor, the data directly to the plurality of models.

2. The system of claim 1, further comprising receiving, by the processor using the data management server, the data from a plurality of data sources, wherein the plurality of data sources comprise a financial risk management system and at least one of:

an authorization system, a demographic system, or a spend/bill system.

3. The system of claim 2, further comprising:

storing, by the processor, the customer variables in a plurality of split files based upon a frequency of use of one or more of the customer variables in the plurality of models; and

reducing, by the processor and based on the storing, a cycle time associated with a repeated execution of the plurality of models.

4. The system of claim 3, wherein the converting uses at least one of a data extraction tool or a Structured Query Language (SQL) macro.

5. The system of claim 4, further comprising standardizing, by the processor using the data management server, the data stored in the SAS flat files.

6. The system of claim 5, wherein the extracting is completed by the data extraction tool, and wherein the data extraction tool includes a user interface, the Structured Query Language (SQL) macro, privacy function, filtering function, and a model building function.

7. The system of claim 6, further comprising validating, by the processor using a model validation tool, the plurality of models.

8. The system of claim 7, wherein the SAS flat file comprises a Statistical Analysis System (SAS) data set.

9. The system of claim 8, further comprising:

scoring, by the processor using the model validation tool, an analytical environment data set;

scoring, by the processor using the model validation tool, a production environment data set; and

comparing, by the processor using the model validation tool, scores generated by the analytical environment data set to scores generated by the production environment data set.

10. The system of claim 9, wherein the results of the comparison between scores generated by the analytical environment data set and scores generated by the production environment are produced in a report.

11. The system of claim 10, further comprising validating, by the processor using the model validation tool, the plurality of models at least in part by using a pre-defined test case.

12. The system of claim 11, further comprising determining, by the processor and using a configuration file, a subset of the plurality of models that have used data elements from a feed, breaking the feed into multiple data files to form split files on production and dynamically generating metadata based on a number of the split files and the customer variables residing on each file.

13. The system of claim 12, wherein the data is gathered by the data management server.

14. The system of claim 13, wherein the customer variables are in the form of SAS datasets.

15. The system of claim 14, further comprising scoring, by the processor and based on the plurality of models, the customers at a scoring frequency.

16. The system of claim 15, wherein the scoring frequency further comprises an expiry date.

17. The system of claim 16, further comprising calling a model of the plurality of models to load in real-time.

18. The system of claim 17, further comprising setting an expiry date function in the model of the plurality of models to specify a length of time for the model to run in a production environment.

19. A method comprising:

converting, by a computer-based system using an extraction tool having a processing Structured Query Language (SQL) macro, Structured Query Language (SQL) variables to Statistical Analysis System (SAS) variables with standardized and consistent Statistical Analysis System (SAS) variable names in SAS flat files on a data management server;

extracting, by the computer-based system, data from the SAS flat files on the data management server based at least in part on contents of customer files respectively associated with customers, wherein the data comprises customer variables,

analyzing, by the computer-based system, the customer variables in the customer files to determine trends in customer behavior;

determining, by the computer-based system, the customer variables to use in a plurality of models, wherein the customer variables referenced in a variable file are determined by the plurality of models that use a particular variable;

determining, by the computer-based system, model logic based on the customer variables and the trends in customer behavior;

enabling, by the computer-based system, creation of the plurality of models based on the model logic; and

delivering, by the computer-based system, the data directly to the plurality of models.

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

converting, by the computer-based system using an extraction tool having a processing Structured Query Language (SQL) macro, Structured Query Language (SQL) variables to Statistical Analysis System (SAS) variables with standardized and consistent Statistical Analysis System (SAS) variable names in SAS flat files on a data management server;

extracting, by the computer-based system, data from the SAS flat files on the data management server based at least in part on contents of customer files respectively associated with customers, wherein the data comprises customer variables,

analyzing, by the computer-based system, the customer variables in the customer files to determine trends in customer behavior;

determining, by the computer-based system, the customer variables to use in a plurality of models, wherein the customer variables referenced in a variable file are determined by the plurality of models that use a particular variable;

determining, by the computer-based system, model logic based on the customer variables and the trends in customer behavior;

enabling, by the computer-based system, creation of the plurality of models based on the model logic; and

delivering, by the computer-based system, the data directly to the plurality of models.