Computing prediction results

• US 20050154687A1
• Filed: 01/14/2004
• Published: 07/14/2005
• Est. Priority Date: 01/14/2004
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Kraiss, Achim
• Granted Patent: US 7,848,909 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models

Abstract

One embodiment of the invention provides a computer-implemented method for providing prediction results to an application system during an interactive session with a user. In this embodiment, the method includes receiving a first set of input values from the application system, using a data mining model along with the first set of input values to compute a first prediction result, and saving state information generated from the computation of the first prediction result. The method further includes receiving a second set of input values from the application system, and using the data mining model along with the state information and the second set of input values to compute a second prediction result.

Claims

1. A computer-implemented method for providing prediction results to an application system during an interactive session with a user, the method comprising:

receiving a first set of input values from the application system;

using a data mining model along with the first set of input values to compute a first prediction result;

saving state information generated from the computation of the first prediction result;

receiving a second set of input values from the application system; and

using the data mining model along with the state information and the second set of input values to compute a second prediction result.

2. The computer-implemented method of claim 1, wherein the method comprises:

sending the first prediction result back to the application system; and

sending the second prediction result back to the application system.

3. The computer-implemented method of claim 1, wherein the second set of input values includes both the first set of input values and an additional set of input values, and wherein the method comprises using the data mining model along with the state information and the additional set of input values to compute the second prediction result.

4. The computer-implemented method of claim 1, wherein the method comprises receiving the first set of input values from the application system before the second set of input values are available.

5. The computer-implemented method of claim 1, wherein the first set of input values includes at least two input values.

6. The computer-implemented method of claim 1, wherein the second set of input values includes at least two input values.

7. The computer-implemented method of claim 1, wherein the method comprises:

receiving the first set of input values from the application system during an interactive session with a customer; and

receiving the second set of input values from the application system during the interactive session with the customer.

8. The computer-implemented method of claim 1, wherein the data mining model is a decision tree model.

9. The computer-implemented method of claim 8, wherein the state information includes information about a particular node in the decision tree model.

10. The computer-implemented method of claim 1, wherein the data mining model is a Naive Bayes model.

11. The computer-implemented method of claim 10, wherein the state information includes intermediate probability information.

12. The computer-implemented method of claim 1, wherein the first and second prediction results each specify a probability of customer churn.

13. A computer system for providing prediction results to an application system during an interactive session with a user, wherein the computer system is programmed to:

receive a first set of input values from the application system;

use a data mining model along with the first set of input values to compute a first prediction result;

save state information generated from the computation of the first prediction result;

receive a second set of input values from the application system; and

use the data mining model along with the state information and the second set of input values to compute a second prediction result.

14. A computer system, comprising:

a data mining model; and

a prediction engine that is operable to;

receive a first input value from an application system;

use the data mining model along with the first input value to compute a first prediction result;

save state information generated from the computation of the first prediction result;

receive a second input value from the application system; and

use the data mining model, the state information, the first input value, and the second input value to compute a second prediction result.

15. The computer system of claim 14, wherein the prediction engine is operable to:

send the first prediction result back to the application system; and

send the second prediction result back to the application system.

16. The computer system of claim 14, wherein the prediction engine is operable to receive the first input value from the application system before the second input value is available.

17. The computer system of claim 14, wherein the data mining model is a decision tree model.

18. The computer system of claim 17, wherein the state information includes information about a particular node in the decision tree model.

19. The computer system of claim 14, wherein the data mining model is a Naive Bayes model.

20. The computer system of claim 19, wherein the state information includes intermediate probability information.

21. A computer-readable medium having computer-executable instructions contained therein for performing a method, the method comprising:

receiving a first set of input values from the application system;

using a data mining model along with the first set of input values to compute a first prediction result;

saving state information generated from the computation of the first prediction result;

receiving a second set of input values from the application system; and

using the data mining model along with the state information and the second set of input values to compute a second prediction result.