Predictive model management using a re-entrant process
View Patent ↗Models are generated using a variety of tools and features of a model generation platform. For example, in connection with a project in which a user generates a predictive model based on historical data about a system being modeled, the user is provided through a graphical user interface a structured sequence of model generation activities to be followed, the sequence including dimension reduction, model generation, model process validation, and model re-generation. A user generates a predictive model based on historical data about a system being modeled. Structured project information is automatically stored that captures a state of the project at successive steps in generating the model.
1. A machine-based method comprising:
in connection with a predictive model development project in which a user interacts with a computer application in a succession of steps to generate a predictive model based on historical data about a system being modeled, the user's progress in developing the model having a state at each of the successive steps,
automatically storing structured project information that captures a state of the project at each of the successive steps in generating the model; the successive steps of the project being re-entrant based on the stored structured project information so that the user can make revisions to the project without restarting the project.
2. The method of claim 1 in which the system comprises behavior of prospective customers of a vendor with respect to a product or service offered by the vendor; behavior of prospective customers belonging to a population of potential customers with respect to a product or service; or behavior of current customers with respect to a current product or service.
3. The method of claim 1 in which the predictive model predicts behavior of a prospective or current customer with respect to purchase of or payment for a product or service of a vendor.
4. The method of claim 1 in which the predictive model predicts behavior of a current customer with respect to retention of a current service or product of a vendor.
5. The method of claim 1 in which the predictive model predicts behavior of a current customer with respect to risk of asserting claims, loan payment or prepayment to a vendor.
6. The method of claim 1 in which the predictive model predicts behavior of a current customer with respect to usage of a current service or product of a vendor.
7. The method of claim 1 in which the project information comprises model process validation and at least two of: project objectives, project schedules, project requirements, information about the historical data, model ensembles and outputs of the model.
8. The method of claim 1 also including enabling the user to replicate a project based on the stored structured project information.
9. The method of claim 1 also including enabling the user to refine a previous project based on the stored structured project information.
10. The method of claim 1 also including enabling the user to apply the model generated in the project based on the stored structured project information.
11. A machine-based method comprising
in connection with a project in which a user generates a predictive model based on historical data about a system being modeled, storing in a common location project information that includes a model validation process and at least two of: project objectives, project schedules, project requirements, information about the historical data, equations expressing the model, performance characteristics of the model, and outputs of the model, steps in generating the model being re-entrant based on the stored project information so that the user can make revisions to the project without restarting the project.
12. The method of claim 11 in which the system comprises behavior of customers of a vendor with respect to products offered by the vendor.
13. The method of claim 11 in which the predictive model predicts behavior of a prospective or current customer with respect to purchase of a product or service of a vendor.
14. The method of claim 11 in which the predictive model predicts behavior of a current customer with respect to retention of a current service or product of a vendor.
15. The method of claim 11 in which the predictive model predicts behavior of a current customer with respect to risk of asserting claims, loan payment or prepayment to a vendor.
16. The method of claim 11 in which the predictive model predicts behavior of a current customer with respect to usage of a current service or product of a vendor.
17. The method of claim 11 in which the project information comprises model process validation and at least two of: project objectives, project schedules, project requirements, information about the historical data, model ensembles and outputs of the model.
18. The method of claim 11 in which the common location comprises a file or folder maintained by an operating system of a computer.
19. The method of claim 11 also including enabling the user to refine a previous project based on the stored structured project information.
20. A machine-based method comprising
enabling users to engage in predictive model development projects to generate predictive models based on historical data about systems being modeled, and
applying a common project tracking paradigm to manage the generation of the models by the users and to store project progress tracking information associated with the respective models in a common format, steps in each model generation project being re-entrant based on the project tracking paradigm so that the users can make revisions to the project without restarting the project.
21. The method of claim 20 in which the predictive models each predicts behavior of a prospective or current customer with respect to purchase of a product or service of a vendor.
22. The method of claim 20 in which the predictive models each predicts behavior of a current customer with respect to retention of a current service or product of a vendor.
23. The method of claim 20 in which the predictive model predicts behavior of a current customer with respect to risk of asserting claims, loan payment or prepayment to a vendor.
24. The method of claim 20 in which the predictive model predicts behavior of a current customer with respect to usage of a current service or product of a vendor.
25. The method of claim 20 in which the project information comprises model process validation and at least two of: project objectives, project schedules, project requirements, information about the historical data, model ensembles and outputs of the model.
26. The method of claim 20 also including enabling the users to replicate a project based on the stored structured project information.
27. The method of claim 20 also including enabling the users to refine a previous project based on the stored structured project information.