Feedback-driven exogenous factor learning in time series forecasting

1. A system for forecast modeling, the system comprising:

at least one processor;

at least one database that is operably coupled to the at least one processor, wherein the database comprises:

a time series data module that is configured to store time series data for a domain,

an exogenous data module that is configured to store exogenous data associated with a plurality of exogenous factors, and

a feedback module that is configured to collect and store feedback data from a plurality of online users, wherein the feedback data is related to the exogenous data and the plurality of exogenous factors;

a data pre-processor module that is configured to use the at least one processor to identify and select a portion of the plurality of exogenous factors using the feedback data collected from the online users for use in a forecast model in combination with the time series data for the domain; and

an auto regressive exogenous (ARX) model module that is operably coupled to the at least one processor and that is configured to create an ARX model and to use the at least one processor to represent coefficients of the selected portion of exogenous factors as a function of input variables in the ARX model.

2. The system of claim 1 wherein:

the feedback data includes positive feedback data and negative feedback data for each of the plurality of exogenous factors; and

the data pre-processor module is configured to select the portion of the plurality of exogenous factors for use in the forecast model when a ratio of positive feedback to negative feedback for an exogenous factor exceeds a threshold.

3. The system of claim 1 wherein the at least one processor is configured to send an online questionnaire to the plurality of online users to obtain the feedback data in response to a triggering event.

4. The system of claim 1 wherein the at least one processor is configured to send an online questionnaire to the plurality of online users to obtain the feedback data in response to a period of time related to the domain.

5. The system of claim 1 wherein the ARX model comprises a linear portion and a fuzzy network portion, wherein at least a portion of coefficients of the linear portion is represented by results from the fuzzy network portion.

6. The system of claim 1 wherein the database comprises an in-memory database.

7. A computer-implemented method for executing instructions stored on a non-transitory computer-readable storage medium, the method comprising:

quantizing and normalizing a plurality of exogenous factors;

collecting and storing feedback data from a plurality of online users related to the plurality of exogenous factors; and

selecting a portion of the plurality of exogenous factors using the feedback data collected from the online users for use in a forecast model.

8. The method as in claim 7 wherein:

the feedback data includes positive feedback data and negative feedback data for each of the plurality of exogenous factors; and

selecting the portion of the exogenous factors comprises selecting the portion of the plurality of exogenous factors for use in the forecast model when a ratio of positive feedback to negative feedback for an exogenous factor exceeds a threshold.

9. The method as in claim 7 further comprising sending an online questionnaire to the plurality of online users to obtain the feedback data in response to a triggering event.

10. The method as in claim 7 further comprising sending an online questionnaire to the plurality of online users to obtain the feedback data in response to a period of time related to the domain.

11. The method as in claim 7 wherein the forecast model includes an auto regressive exogenous (ARX) model.

12. The method as in claim 11 wherein the ARX model comprises a linear portion and a fuzzy network portion, wherein at least a portion of coefficients of the linear portion is represented by results from the fuzzy network portion.

13. The method as in claim 11 further comprising representing coefficients of the selected portion of exogenous factors as a function of input variables in the ARX model.

14. The method as in claim 13 further comprising predicting a future time series data using historical time series data for a domain and the selected portion of the exogenous factors in the ARX model.

15. A computer program product, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to:

quantize and normalize a plurality of exogenous factors;

collect and store feedback data from a plurality of online users related to the plurality of exogenous factors; and

select a portion of the plurality of exogenous factors using the feedback data collected from the online users for use in a forecast model.

16. The computer program product of claim 15 wherein the forecast model includes an auto regressive exogenous (ARX) model.

17. The computer program product of claim 16 wherein the ARX model comprises a linear portion and a fuzzy network portion, wherein at least a portion of coefficients of the linear portion is represented by results from the fuzzy network portion.

18. The computer program product of claim 16 further comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to represent coefficients of the selected portion of exogenous factors as a function of input variables in the ARX model.

19. The computer program product of claim 18 further comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to predict a future time series data using historical time series data for a domain and the selected portion of the exogenous factors in the ARX model.