Process for determining a confidence factor for insurance underwriting suitable for use by an automated system

• US 20030187698A1
• Filed: 06/17/2002
• Published: 10/02/2003
• Est. Priority Date: 12/31/2001
• Status: Active Grant

• Current Assignee: Pacific Life Insurance Company (Pacific Mutual Holding Company)
• Original Assignee: Genworth Financial Incorporated
• Inventors: Bonissone, Piero Patrone, Cheetham, William Estel
• Granted Patent: US 8,005,693 B2
• Time in Patent Office: Days
• US Class Current: 705/4
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

A process is described for evaluating the decision-making confidence of a process and system for at least a partial underwriting of insurance policies where placement of an insurance application to an underwriting category is based on its similarity to previous insurance applications. The confidence factor computed is a measure of the correctness of the decision for a given application for insurance.

Claims

1. A process for determining the confidence factor for an insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

estimating the conditional probability of misclassification for the at least one identified parameter;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

defining a run-time function to evaluate the confidence threshold based on the soft constraint.

2. The process according to claim 1, where at least one identified parameter comprises at least one of:

a) the number of retrieved cases;

b) the variability of the retrieved cases;

c) the number of retrieved cases thresholded by similarity value;

d) the variability of retrieved cases thresholded by similarity value

e) the number of refined cases;

f) the variability of refined cases

g) the number of refined cases, thresholded by similarity value

h) the variability of refined cases thresholded by similarity value

i) the measure of strength of mode;

j) the number of retrieved cases weighted by similarities;

k) the variability of retrieved cases weighted by similarities;

1) the number of refined cases weighted by similarities; and

m) the variability of refined cases weighted for similarities

3. The process according to claim 1, wherein the step of estimating the conditional probability of misclassification for the at least one identified parameter further comprises:

accessing a case base containing a plurality of cases whose associated application decisions have been certified correct;

selecting one of the plurality of cases, where the selected case is defined as the probe case;

identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

comparing the underwriting classification determination with the original certified decision of the probe case; and

recording the comparison and at least one identified internal parameter for the probe case.

4. The process according to claim 1, wherein the step of translating the conditional probability of misclassification into a soft constraint for each parameter further comprises a penalty for misclassification and a reward for correct classification.

5. The process according to claim 1, wherein the step of translating the conditional probability of misclassification into a soft constraint for each parameter further comprises an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications.

6. The process according to claim 1, further comprising the step generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

7. The process according to claim 6, further comprising the step of computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.

8. The process according to claim 1, wherein the insurance underwriting decision is an automated decision.

9. A process for determining the confidence factor for an application decision based on a comparison of at least one previous application decision, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the application;

estimating the conditional probability of misclassification for the at least one identified parameter;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

defining a run-time function to evaluate the confidence threshold based on the soft constraint.

10. The process according to claim 9, wherein the step of estimating the conditional probability of misclassification for the at least one identified parameter further comprises:

accessing a case base containing a plurality of cases whose associated application decisions have been certified correct;

selecting one of the plurality of cases, where the selected case is defined as the probe case;

identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

comparing the underwriting classification determination with the original certified decision of the probe case; and

recording the comparison and the at least one identified internal parameter for the probe case.

11. The process according to claim 9, wherein the step of translating the conditional probability of misclassification into a soft constraint for each parameter further comprises a penalty for misclassification and a reward for correct classification.

12. The process according to claim 9, wherein the step of translating the conditional probability of misclassification into a soft constraint for each parameter further comprises an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications.

13. The process according to claim 9, further comprising the step generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

14. The process according to claim 13, further comprising the step of computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.

15. The process according to claim 9, wherein the application decision is an automated application decision.

16. A process for determining the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

estimating the conditional probability of misclassification for the at least one identified parameter, including the sub-steps of;

a) accessing a case base containing a plurality of cases whose associated application decisions have been certified correct;

b) selecting one of the plurality of cases, where the selected case is defined as the probe case;

c) identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

d) determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

e) comparing the underwriting classification determination with the original certified decision of the probe case; and

f) recording the comparison and the at least one identified internal parameter for the probe case;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

defining a run-time function to evaluate the confidence threshold based on the soft constraint.

17. A process for determining the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

estimating the conditional probability of misclassification for the at least one identified parameter;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter, including comprising a penalty for misclassification and a reward for correct classification; and

defining a run-time function to evaluate the confidence threshold based on the soft constraint.

18. A process for determining the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

estimating the conditional probability of misclassification for the at least one identified parameter;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter through an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications; and

defining a run-time function to evaluate the confidence threshold based on the soft constraint.

19. A process for determining the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the process comprising:

identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

estimating the conditional probability of misclassification for the at least one identified parameter;

translating the conditional probability of misclassification into a soft constraint for the at least one parameter;

defining a run-time function to evaluate the confidence threshold based on the soft constraint; and

generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

20. The process according to claim 19, further comprising the step of computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.

21. A medium storing code for causing a processor to determine the confidence factor for an insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

code for estimating the conditional probability of misclassification for the at least one identified parameter;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

code for defining a run-time function to evaluate the confidence threshold for each new query.

22. The medium according to claim 21, where the at least one identified parameters comprise at least one of:

a) the number of retrieved cases;

b) the variability of the retrieved cases;

c) the number of retrieved cases thresholded by similarity value;

d) the variability of retrieved cases thresholded by similarity value

e) the number of refined cases;

f) the variability of refined cases

g) the number of refined cases, thresholded by similarity value

h) the variability of refined cases thresholded by similarity value

i) the measure of strength of mode;

j) the number of retrieved cases weighted by similarities;

k) the variability of retrieved cases weighted by similarities;

l) the number of refined cases weighted by similarities; and

m) the variability of refined cases weighted for similarities

23. The medium according to claim 21, wherein the code for estimating the conditional probability of misclassification for the at least one identified parameters further comprises:

code for accessing a case base containing a plurality of cases whose associated application decisions have been certified correct;

code for selecting one of the plurality of cases, where the selected case is defined as the probe case;

code for identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

code for determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

code for comparing the underwriting classification determination with the original certified decision of the probe case; and

code for recording the comparison and the at least one identified internal parameter for the probe case.

24. The medium according to claim 21, wherein translating the conditional probability of misclassification into a soft constraint for each parameter further comprises a penalty for misclassification and a reward for correct classification.

25. The medium according to claim 21, wherein translating the conditional probability of misclassification into a soft constraint for each parameter further comprises an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications.

26. The medium according to claim 21, further comprising code for generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

27. The medium according to claim 26, further comprising code for computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.

28. The medium according to claim 21, wherein the insurance underwriting decision is an automated decision.

29. A medium storing code for causing a processor to determining the confidence factor for an application decision based on a comparison of at least one previous application decision, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the application;

code for estimating the conditional probability of misclassification for the at least one identified parameters;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

code for defining a run-time function to evaluate the confidence threshold based on the soft constraint.

30. The medium according to claim 29, wherein the code for estimating the conditional probability of misclassification for the at least one identified parameter further comprises:

code for accessing a case base containing a plurality of cases whose associated application decisions have been were certified correct;

code for selecting one of the plurality of cases, where the selected case is defined as the probe case;

code for identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

code for determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

code for comparing the underwriting classification determination with the original certified decision of the probe case; and

code for recording the comparison and the at least one identified internal parameter for the probe case.

31. The medium according to claim 29, wherein translating the conditional probability of misclassification into a soft constraint for each parameter further comprises a penalty for misclassification and a reward for correct classification.

32. The medium according to claim 29, wherein translating the conditional probability of misclassification into a soft constraint for each parameter further comprises an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications.

33. The medium according to claim 29, further comprising code for generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

34. The medium according to claim 29, further comprising code for computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.

35. The medium according to claim 29, wherein the application decision is an automated application decision.

36. A medium storing code for causing a processor to determine the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

code for estimating the conditional probability of misclassification for the at least one identified parameter, including;

a) code for accessing a case base containing a plurality of cases whose associated insurance application decisions have been certified correct;

b) code for selecting one of the plurality of cases, where the selected case is defined as the probe case;

c) code for identifying at least one internal parameter of the probe case that may affect the probability of misclassification;

d) code for determining the underwriting classification of the probe case based on the remaining plurality of cases within the case base;

e) code for comparing the underwriting classification determination with the original certified decision of the probe case; and

f) code for recording the comparison and the at least one identified internal parameter for the probe case;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter; and

code for defining a run-time function to evaluate the confidence threshold based on the soft constraint.

37. A medium storing code for causing a processor to determine the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

code for estimating the conditional probability of misclassification for the at least one identified parameter;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter, including comprising a penalty for misclassification and a reward for correct classification; and

code for defining a run-time function to evaluate the confidence threshold for based on the soft constraint.

38. A medium storing code for causing a processor to determine the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

code for estimating the conditional probability of misclassification for the at least one identified parameter;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter through an aggregating functions using a weighted sum of rewards and penalties with increasing penalties for misclassifications; and

code for defining a run-time function to evaluate the confidence threshold for based on the soft constraint.

39. A medium storing code for causing a processor to determine the confidence factor for an automated insurance application underwriting decision based on a comparison of at least one previous underwritten insurance application, the medium comprising:

code for identifying at least one internal parameter that may affect the probability of misclassification of the insurance application;

code for estimating the conditional probability of misclassification for the at least one identified parameter;

code for translating the conditional probability of misclassification into a soft constraint for the at least one parameter;

code for defining a run-time function to evaluate the confidence threshold based on the soft constraint; and

code for generating a soft constraint evaluation vector that contains the degree to which each parameter satisfies its corresponding soft constraint.

40. The medium according to claim 39, further comprising code for computing a confidence factor based on the intersection of all the soft constraints evaluations contained in the soft constraint evaluation vector.