System And Process For Dominance Classification For Insurance Underwriting Suitable For Use By An Automated System
A risk classification technique that exploits the existing risk structure of the decision problem in order to produce risk categorizations for new candidates is described. The technique makes use of a set of candidates for which risk categories have already been assigned (in the case of insurance underwriting, for example, this would pertain to the premium class assigned to an application). Using this set of labeled candidates, the technique produces two subsets for each risk category: the Pareto-best subset and the Pareto-worst subset by using Dominance. These two subsets can be seen as representing the least risky and the most risky candidates within a given risk category. If there are a sufficient number of candidates in these two subsets, then the candidates in these two subsets can be seen as samples from the two hypothetical risk surfaces in the feature space that bound the risk category from above and below respectively. A new candidate is assigned a risk category by verifying if the candidate lies within these two bounding risk surfaces.
1 - 41 . (canceled)
42 . A process performed by a computing device for underwriting an insurance application based on a plurality of previous insurance application underwriting decisions, the process comprising:
gathering, by the computing device, a plurality of previous insurance application underwriting decisions, where the plurality of decisions have resulted in the same classification for respective prior applications to which the decisions are associated with, the plurality of decisions each comprise a plurality of features;
receiving, by the computing device, a request to underwrite an insurance application, where the request includes information about at least one application component;
defining, by the computing device, a first set comprising at least one of the plurality of previous insurance application underwriting decisions that is not dominated by any other of the plurality of previous insurance application underwriting decisions, such that the first set constitutes a first subset;
defining, by the computing device, a second set comprising at least one of the plurality of previous insurance application underwriting decisions that does not dominate any of the other of the plurality of previous insurance application underwriting decisions, such that the second set constitutes a second subset, the first set and the second set being constituted by different previous insurance application underwriting decisions;
the first set and the second set associated with respective prior applications; and
comparing, by the computing device, the insurance application to the first set and the second set; and
wherein the insurance application comprises a plurality of features; and
where the comparing the insurance application to the first set and the second set further comprises:
comparing the at least one feature to a corresponding feature in the at least one of the plurality of previous insurance application underwriting decisions in the first set; and
comparing the at least one feature to a corresponding feature in the at least one of the plurality of previous insurance application underwriting decisions in the second set; and
where:
comparing the at least one feature to a corresponding feature in the first set further comprises determining if the at least one feature in the insurance application is dominated by the corresponding feature in the first set to generate a first dominated determination; and
comparing the at least one feature to a corresponding feature in the second set further comprises determining if the at least one feature in the insurance application is dominated by the corresponding feature in the second set to generate a second dominated determination;
each of the comparings including determining whether each of said feature in the insurance application represents greater or less risk than the corresponding feature in the first set or second set, respectively;
the process further including, based on the first and second dominated determinations, assessing, by the computing device, whether all the features of the application are dominated by all the corresponding features of the first set, and including assessing whether all the features of the application are dominated by all the corresponding features of the second set;
the process further including assigning, by the computing device, a risk assessment to the insurance application based on the assessings; and
outputting, by the computing device, the risk assessment.
43 . The system according to claim 42 , further comprising an assignment module for assigning a classification to the insurance application based on the risk assessment.
44 . The system of claim 42 , wherein the first subset is constituted by a Pareto best subset.
45 . The system of claim 42 , wherein the second subset is constituted by a Pareto worst subset.
46 . A process performed by a computing device for underwriting an insurance application based on a plurality of previous insurance application underwriting decisions, the process comprising:
gathering, by the computing device, a plurality of previous insurance application underwriting decisions, where the plurality of decisions have resulted in the same classification for respective prior applications to which the decisions are associated with, the plurality of decisions each comprise a plurality of features;
receiving, by the computing device, a request to underwrite an insurance application, where the request includes information about at least one application component;
defining, by the computing device, a first set comprising at least one of the plurality of previous insurance application underwriting decisions that is not dominated by any other of the plurality of previous insurance application underwriting decisions, such that the first set constitutes a first subset;
defining, by the computing device, a second set comprising at least one of the plurality of previous insurance application underwriting decisions that does not dominate any of the other of the plurality of previous insurance application underwriting decisions, such that the second set constitutes a second subset, the first set and the second set being constituted by different previous insurance application underwriting decisions;
the first set and the second set associated with respective prior applications; and
comparing, by the computing device, the insurance application to the first set and the second set; and
wherein the insurance application comprises a plurality of features; and
where the comparing the insurance application to the first set and the second set further comprises:
comparing the at least one feature to a corresponding feature in the at least one of the plurality of previous insurance application underwriting decisions in the first set; and
comparing the at least one feature to a corresponding feature in the at least one of the plurality of previous insurance application underwriting decisions in the second set; and
each of the comparings including determining whether each of said feature in the insurance application represents greater or less risk than the corresponding feature in the first set or second set, respectively;
the process further including, based on the first and second dominated determinations, assessing, by the computing device, whether all the features of the application are dominated by all the corresponding features of the first set, and including assessing whether all the features of the application are dominated by all the corresponding features of the second set;
the process further including assigning, by the computing device, a risk assessment to the insurance application based on the assessings; and
outputting, by the computing device, the risk assessment.