INTERFACES FOR PREDICTIVE MODELS

• US 20140245207A1
• Filed: 03/15/2013
• Published: 08/28/2014
• Est. Priority Date: 02/25/2013
• Status: Active Grant

• Current Assignee: Christian D. Poulin
• Original Assignee: Christian D. Poulin
• Inventors: Poulin, Christian D.
• Granted Patent: US 9,235,332 B2
• Time in Patent Office: Days
• US Class Current: 715/771
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G06F 18/2111 by using evolutionary computational techniques, e.g. genetic algorithms
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/2323 based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06F 3/0484 for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

User interfaces for tools for estimating a probability that a future event will occur based on user input are described. One set of interfaces include rating, trend, cohort record and source controls each of which when selected provide corresponding data from one predictive model that produces predictions of the likelihood of an event occurring in the future based on analysis of data in a database. The system further displays a process of content produced by a model builder that populates the interfaces, and outputs thereof.

Claims

1. A method executed in a computer system for providing a user interface to a tool for estimating a probability that a future event will occur based on user input, the method comprises:

generating by one or more computer systems a first set of interfaces that include rating, trend, cohort, record, and source controls each of which when selected provide corresponding data from one predictive model that produces predictions of the likelihood of an event occurring in the future based on analysis of data in a database; and

sending by the by one or more computer systems the first set of interface outputs to a user system.

2. The method of claim 1, wherein the generated interfaces has a mode control to select between a first basic modes and a second enhanced mode, with the basic mode, users are presented with an overall statistical rating, with supporting statistics, as well as analyst notes, and with the enhanced mode users further are able to build models.

3. The method of claim 1, wherein the generated interfaces include an initial search interface comprising a search bar that renders found published predictive models meeting search criteria.

4. The method of claim 1, wherein the generated interfaces users can select according to model type or by model topic.

5. The method of claim 1, wherein the generated interfaces include a region to render search result for a model, with a determined prediction.

6. The method of claim 1, wherein the generated interfaces include a region to render search results for model, with determined predictions with for each result returned for a found model controls are also returned corresponding to the rating control, the trend control, the by cohort control, the record control, and the source control.

7. The method of claim 1, wherein the rating control launches the software to determine a prediction of risk on a given day or time.

8. The method of claim 1, wherein the trend control produces a chart of predictions for the security vs. the actual security values.

9. The method of claim 1, wherein the “

by cohort”

control provides a comparison.

10. The method of claim 1, wherein the “

the record icon provides an area for analyst annotated notes on the model and the source icon provides content from a source.

11. The method of claim 1, wherein the first set of interfaces includes a results area which displays plural results of found models, with each result having a dedicated set of the rating, trend, cohort record and source controls for a model corresponding to the result.

12. The method of claim 1, wherein the results area for a particular model further includes a signal strength value wherein tested accuracy of the model is displayed.

13. A method executed in a computer system for providing a user interface to model for estimating a probability that a future event will occur based on user input, the method comprises:

generating by one or more computer systems a first set of interfaces that include data source, model, test case and prediction tabs each of which when selected provide corresponding data from one predictive model that produces predictions of the likelihood of an event occurring in the future based on analysis of data in a database; and

sending by the by one or more computer systems the first set of interface outputs to a user system.

14. The method of claim 13 wherein when the models tab is selected the tab allows the user to create a model or sort models by date, and filter models according to privileges, e.g., private, public or all.

15. The method of claim 13 wherein when the data sources tab is selected an overlay window is provided to aid the user in selecting data sources from which to collect data that will be used in building the model.

16. The method of claim 13, wherein when the test cases tab is selected the interface presents the user an overlay window with fields to select a model name an option description and a sharing level, e.g., private or public and a control to apply a test case to test the selected model.

17. The method of claim 13, wherein when the predictions tab is selected predictions are rendered from execution of the model and a sort control sorts predictions according to date and by model performance.

18. A system for providing a user interface to a tool for estimating a probability that a future event will occur based on user input comprises:

a processor;

memory operatively connected to the processor; and

a computer storage device storing a computer program product for generating by the processor a first set of interfaces, the computer program product comprising instructions for causing the processor to

generate the first set of interfaces, with the first set of interfaces including rating, trend, cohort, record, and source controls each of which when selected provide corresponding data from one predictive model that produces predictions of the likelihood of an event occurring in the future based on analysis of data in a database; and

send the first set of interfaces to a user system.

19. The system of claim 18, wherein the first set of interfaces includes a results area which displays plural results of found models, with each result having a dedicated set of the rating, trend, cohort record and source controls for a model corresponding to the result.

20. A computer program product tangibly stored on a computer storage device the computer program product for providing a user interface to a tool for estimating a probability that a future event will occur based on user input, the computer program product comprises instructions to:

generate the first set of interfaces, with the first set of interfaces including rating, trend, cohort, record, and source controls each of which when selected provide corresponding data from one predictive model that produces predictions of the likelihood of an event occurring in the future based on analysis of data in a database; and

send the first set of interface outputs to a user system.

21. The computer program product of claim 20, wherein the first set of interfaces includes a results area which displays plural results of found models, with each result having a dedicated set of the rating, trend, cohort record and source controls for a model corresponding to the result.