Insurance pattern analysis

• US 7,398,218 B1
• Filed: 08/28/2000
• Issued: 07/08/2008
• Est. Priority Date: 08/27/1999
• Status: Expired due to Term

• Current Assignee: DUCK Creek Technologies Incorporated
• Original Assignee: Accenture LLP (Accenture PLC)
• Inventors: Thompson, Eric W., Krans, Michael R., Famulla, Rainer M., Bernaski, Michael J., Davis, Nancy J., DeVoie, Kara J., Freeland, John G., Albright, Jerome J., Bolton, James R.
• Primary Examiner(s): Frenel; Vanel
• Application Number:
  US09/649,766
• Time in Patent Office: 2,871 Days
• Field of Search: 705 2- 4, 705/10, 705/37, 705/80
• US Class Current: 705/3
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

A method of determining risk includes comparing case data with a collection of profiles, selecting profiles from the collection of profiles, and aggregating the outcomes associated with the selected profiles to determine the risk associated with the case data.

Claims

1. A method of determining risk, comprising:

a computer segmenting case data of a single insurable customer into a plurality of components associated with a collection of profiles;

the computer selecting a plurality of profiles from the collection of profiles that match the components associated with the collection of profiles, each of the selected profiles being representative of a respective part of the case data and an insurable dimension of the single insurable customer;

selecting with the computer at least one of a plurality of predefined qualities for each of the components in each one of the selected profiles based on application of quality definitions to the case data segmented into each one of the respective components, the quality definitions associated with each of the selected profiles and each of the selected at least one of a plurality of predefined qualities is a term that provides a non-numeric valuation that is descriptive of the case data that is segmented into a respective one of the components within a respective one of the profiles;

for each one of the selected profiles generating with the computer an outcome by logical combination of the predefined qualities selected for each of the components in each one of the selected profiles, the outcome of a selected profile is a pre-defined expression that is descriptive of the respective part of the case data and the insurable dimension of the single insurable customer represented with the selected profile; and

the computer aggregating the outcomes from each of the selected profiles to determine the risk associated with the case data of the single insurable customer.

2. The method of claim 1, wherein the risk is the underwriting risk associated with policies selected from the group consisting of medical insurance, business owners insurance, commercial automobile insurance, workers compensation insurance, professional liability insurance, umbrella liability insurance, recreational package insurance and life insurance.

3. The method of claim 1, wherein the segmenting comprises:

matching the components to the case data by;

matching the case data with characteristics;

matching the case data and the characteristics with collectives; and

matching the case data and the characteristics and the collectives with collective-collectives.

4. The method of claim 1, further comprising receiving the case data from a network.

5. The method of claim 4, wherein the network is the internet.

6. The method of claim 5, further comprising sending data corresponding to the risk associated with the case data out over the internet.

7. The method of claim 1 further comprising sending data corresponding to the risk associated with the case data out over a network.

8. The method of claim 7, wherein the network is the internet.

9. The method of claim 1, wherein generating an outcome comprises selecting an outcome based on logical aggregation of the predefined qualities selected for each of the components with at least one of a logical AND or a logical OR.

10. The method of claim 1, wherein the risk is underwriting risk of a business owner policy.

11. The method of claim 10, wherein the collection of profiles comprises business risk profiles, property risk profiles and general liability risk profiles.

12. The method of claim 11, wherein the case data comprises data for a plurality of buildings.

13. The method of claim 11, wherein the case data comprises data for a plurality of business locations.

14. The method of claim 12, wherein the aggregating comprises:

aggregating together outcomes of selected property risk profiles, to form an aggregated property risk profile outcome;

aggregating together outcomes of selected general liability risk profiles, to form an aggregated general liability risk profile outcome; and

aggregating together outcomes of selected business risk profiles, the aggregated property risk profile outcome and the aggregated general liability risk profile outcome.

15. The method of claim 13, wherein the aggregating comprises:

aggregating together outcomes of selected property risk profiles, to form an aggregated property risk profile outcome;

aggregating together outcomes of selected general liability risk profiles, to form an aggregated general liability risk profile outcome; and

aggregating together outcomes of selected business risk profiles, the aggregated property risk profile outcome and the aggregated general liability risk profile outcome.

16. A computer program product, comprising:

a computer-readable medium,

means stored on the medium for segmenting case data into a plurality of components associated with a collection of profiles;

means stored on the medium for selecting a plurality of profiles from the collection of profiles that match the components associated with the collection of profiles, each of the selected profiles being representative of a respective part of the case data and an insurable dimension of a single insurable customer;

means stored on the medium for selecting at least one of a plurality of predefined qualities for each one of the components to describe factors included in the case data segmented into each one of the respective components by application of quality definitions to the case data segmented into each one of the respective components, the quality definitions are associated with each of the selected profiles, and the predefined qualities provide a non-numeric valuations each of which is descriptive of the case data segmented into a respective one of the components within a respective one of the profiles;

means stored on the medium for generating an outcome for each of the selected profiles based on logical combination of the predefined qualities that were selected for each one of the respective components within a respective one of the profiles, the outcome for each of the selected profiles is an expression selected from among a group of pre-defined expressions that is descriptive of the respective part of the case data and the insurable dimension of the single insurable customer represented with the selected profile; and

means stored on the medium for aggregating the outcomes of each of the selected profiles to determine a risk associated with the entirety of the case data of the single insurable customer.

17. The computer program product of claim 16, wherein the risk is the underwriting risk associated with policies selected from the group consisting of medical insurance, business owners insurance, commercial automobile insurance, workers compensation insurance, professional liability insurance, umbrella liability insurance, recreational package insurance and life insurance.

18. The computer program product of claim 16, wherein the segmenting comprises:

matching the components to the case data by;

matching the case data with characteristics;

matching the case data and the characteristics with collectives; and

matching the case data, the characteristics and the collectives with collective-collectives.

19. The computer program product of claim 16, wherein the risk is underwriting risk of a business owner policy.

20. The computer program product of claim 19, wherein the collection of profiles comprises business risk profiles, property risk profiles and general liability risk profiles.

21. The computer program product of claim 20, wherein the case data comprises data for a plurality of buildings.

22. The computer program product of claim 20, wherein the case data comprises data for a plurality of business locations.

23. The computer program product of claim 21, wherein the aggregating comprises:

aggregating together outcomes of selected property risk profiles, to form an aggregated property risk profile outcome;

aggregating together outcomes of selected general liability risk profiles, to form an aggregated general liability risk profile outcome; and

aggregating together outcomes of selected business risk profiles, the aggregated property risk profile outcome and the aggregated general liability risk profile outcome.

24. The computer program product of claim 22, wherein the aggregating comprises:

aggregating together outcomes of selected property risk profiles, to form an aggregated property risk profile outcome;

aggregating together outcomes of selected general liability risk profiles, to form an aggregated general liability risk profile outcome; and

aggregating together outcomes of selected business risk profiles, the aggregated property risk profile outcome and the aggregated general liability risk profile outcome.

25. The computer program product of claim 16, wherein the means stored on the medium for generating an outcome for each of the selected profiles is configured to logically combine the predefined qualities with at least one of a logical AND or a logical OR.

26. A method of evaluating insurance related business, the method comprising:

a computer segmenting data of a single insurable customer into a plurality of components that are associated with a plurality of profiles;

the computer selecting profiles from among the plurality of profiles that match the components so that each of the selected profiles are associated with one of a plurality of different insurable dimensions of the insurance related business;

the computer applying gradient definitions of respective discrete gradients for each one of the components associated with the selected profiles to the respective segmented data;

generating respective discrete gradients that are descriptive of the segmented data for each of the respective components with the computer based on the gradient definitions;

the computer logically combining two or more discrete gradients of the respective components with a logical AND or a logical OR to arrive at a discrete outcome for each of the selected profiles that is indicative of a respective one of the insurable dimensions; and

the computer aggregating the discrete outcomes of each of the selected profiles to obtain a discrete result.

27. The method of claim 26, wherein multiple parts are included in at least one of the insurable dimensions and selecting profiles comprises selecting at least one profile for each of the multiple parts.

28. The method of claim 27, wherein aggregating the discrete outcomes comprises aggregating the discrete outcomes of each of the selected profiles within each of the dimensions to provide an aggregated discrete outcome of each of the dimensions in a first tier.

29. The method of claim 28, wherein aggregating the discrete outcomes comprises aggregating the aggregated discrete outcomes of each of the dimensions to form an aggregated discrete outcome in a second tier that is a score of the insurance related business.

30. The method of claim 28, wherein at least one of the dimensions includes a single part and aggregating the discrete outcome comprises aggregating the aggregated discrete outcomes of each of the dimensions in the first tier with a discrete outcome of at least one of the dimensions that includes a single part to form an aggregated discrete outcome in a second tier.

31. The method of claim 26, wherein applying gradient definitions comprises identifying, with one of the respective discrete gradients, a quality of the respective component associated with one of the profiles.

32. The method of claim 31, wherein identifying a quality comprises assigning a determined description indicative of the respective component.

33. A method of evaluating insurance related business, the method comprising:

a computer segmenting data into a plurality of components associated with a plurality of profiles as a function of the insurance related business to which the data relates;

the computer identifying different profiles that are associated with at least one of the components and are related to different dimensions of the insurance related business;

the computer applying gradient definitions associated with each of the respective components to the data segmented into a respective one of the components;

the computer generating a gradient for each of the respective components, the gradient being a non-numeric quality indication descriptive of the data segmented into each of the respective components;

the computer logically combining the non-numeric quality indications of each of the components in a respective profile to determine a first profile outcome for each of the profiles, the first profile outcome selected from among a group of predefined first profile outcomes to be a first non-numeric valuation of the data associated with the respective profile;

the computer logically combining the first profile outcomes of each of the profiles based on the different dimensions to which each of the profiles relate to determine a second outcome for each of the different dimensions of the insurance related business, wherein the second outcome is selected from among a group of predefined discrete second outcomes to be a second non-numeric valuation that is indicative of one of the dimensions; and

aggregating the second outcomes of each of the dimensions to obtain an aggregated discrete third outcome representative of a third non-numeric valuation of the insurance related business.

34. The method of claim 33, wherein the aggregated discrete third non-numeric outcome is one of a plurality of aggregated discrete third outcomes for each of a plurality of classes identified within the insurance related business and the method further comprises aggregating the aggregated discrete third outcomes to obtain a discrete result selected from among a group of predefined discrete results to be conclusive of the insurance related business.

35. The method of claim 33, wherein the components comprise a collective category having at least one of collective characteristics and collective-collective characteristics in a logical relationship.

36. The method of claim 33, wherein aggregating the second outcomes to obtain an aggregated discrete third outcome comprises logically combining the second outcomes of the different dimensions classified in a class of the insurance related business.

37. The method of claim 33, wherein comparing data with a plurality of profiles comprises comparing data of a specific case to profiles associated with predetermined classes of the insurance related business.

38. The method of claim 33, wherein identifying different profiles comprises matching the profiles in different predetermined classes with the data that has been segmented into the components.

39. The method of claim 33, wherein the computer logically combining the non-numeric quality indications of each of the components in a respective profile comprises the computer logically combining the non-numeric quality indications with at least one of a logical AND or a logical OR.

40. A system for evaluation of insurance related data, the system comprising:

a computer;

a profile database connected to the computer, wherein the profile database includes a collection of profiles each associated with a respective plurality of components each of the components comprising gradient definitions associated with a plurality of respective variables;

an engine executable by the computer to access the profile database, configured to segment insurance related data of a specific case into segments that are each associated with at least one of the components, and match a plurality of profiles from among the collection of profiles to the respective components into which the insurance related data of the specific case has been segmented,

the engine further configured to identify predefined discrete gradients that are each a non-numeric valuation that is descriptive of the segment of the insurance related data associated with one of the components by logical application of the gradient definitions from the respective one of the components within the matched profiles to the segment of the insurance related data associated with the respective one of the components as a function of the respective variables,

the engine further configured to logically combine the discrete gradients of each of the components in a respective one of the matched profiles, and select a discrete outcome for each of the matched profiles from among a group of predefined discrete outcomes,

the engine further configured to aggregate the discrete outcomes of each of the matched profiles to form an aggregated outcome that is selected from among a group of predefined aggregated outcomes.

41. The system of claim 40, wherein the plurality of profiles is representative of one of a plurality of determined classes within the insurance related data and the engine is operable to develop an aggregated outcome for each of the determined classes.

42. The system of claim 41, wherein the engine is operable to aggregate the aggregated outcomes from each of the determined classes to form a conclusive result, whereby the conclusive result is a score that provides an evaluation of the insurance related data.

43. The system of claim 40, wherein the aggregated outcome is representative of one of a plurality of dimensions of the insurance related data.

44. The system of claim 40, wherein at least two profiles utilize a common logical association of variables.

45. The system of claim 40, further comprising a profile builder connected to the computer, wherein the profile builder is operable to create and modify profiles storable in the profile database.

46. The system of claim 45, wherein the profiles each include profile elements, and the profile builder is operable to identify a version within each of at least one of the profiles and the profile elements, wherein the version includes a start date and a stop date indicative of a time frame of applicability.

47. The system of claim 40, further comprising a results database coupled with the computer, wherein the results database is operable to store results generated by the engine.

48. The system of claim 40, wherein the engine is further configured to logically combine the discrete gradients of each of the components in a respective one of the matched profiles with at least one of a logical AND or a logical OR.