Insurance pattern analysis
First Claim
Patent Images
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.
15 Assignments
0 Petitions
Accused Products
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.
-
Citations
48 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. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
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. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25)
-
-
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. - View Dependent Claims (27, 28, 29, 30, 31, 32)
-
-
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. - View Dependent Claims (34, 35, 36, 37, 38, 39)
-
-
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. - View Dependent Claims (41, 42, 43, 44, 45, 46, 47, 48)
-
Specification