IP Library Granted Patent US 7,899,688
Granted Patent B2
US 7,899,688 · App. 10/173,017 · Granted Mar 1, 2011

Process for optimization of insurance underwriting suitable for use by an automated system

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,899,688
App. No.
10/173,017
Granted
Mar 1, 2011
Kind
B2
Abstract

A robust process for automating the tuning and maintenance of decision-making systems is described. A configurable multi-stage mutation-based evolutionary algorithm optimally tunes the decision thresholds and internal parameters of fuzzy rule-based and case-based systems that decide the risk categories of insurance applications. The tunable parameters have a critical impact on the coverage and accuracy of decision-making, and a reliable method to optimally tune these parameters is critical to the quality of decision-making and maintainability of these systems.

Claims (33)

1. A computer implemented process for optimizing an insurance application underwriting process for an insurance underwriting system, the computer implemented process performed by a tangible computer device, the process comprising:

rendering, by the computer device, a first underwriting decision for each insurance application based on respectively applying a first process to the parameters of each insurance application of a plurality of insurance applications, the first underwriting decision comprising a first classification of each insurance application, the first underwriting decisions being designated as included in a case base;

defining a second process for performing underwriting decisioning;

rendering, by the computer device, a second underwriting decision for each insurance application based on respectively applying the second process to the parameters of the each insurance application, the second underwriting decision comprising a second classification of each insurance application, the second process utilizing a plurality of fuzzy rules utilizing decision thresholds;

comparing, by the computer device, the first underwriting decisions to the second underwriting decisions to generate a result, the result relating to a degree of matching between the first underwriting decision rendered for an insurance application and the second underwriting decision rendered for the same application;

outputting, by the computer device, the result of the comparing between the first underwriting decision and the second underwriting decision performed by the process; and

further comprising generating at least one penalty based on the comparing of the second underwriting decisions and the first underwriting decisions, the generating at least one penalty including aggregating (1) penalties for misclassifications, between the second underwriting decisions and the first underwriting decisions, with (2) rewards for correct classifications, between the second underwriting decisions and the first underwriting decisions;

wherein penalties for misclassification are based on a determination of number of rate classes that a second underwriting decision is as compared to a corresponding first underwriting decision;

where the first process and the second process are a plurality of underwriting rules; and

further comprising redefining the plurality of underwriting rules based on the generated penalty and the comparing of the first underwriting decisions and the second underwriting decisions performed by applying the plurality of underwriting rules, the redefining including tuning, by the computer device, the decision thresholds of the fuzzy rules.

2. The computer implemented process according to claim 1 , where the process is an application comparison of an insurance application with at least one previously made insurance application underwriting decision, and the result of the comparison is in the form of a matrix.

3. The computer implemented process according to claim 1 , further comprising the steps of:

performing a plurality of insurance application underwriting decisions by applying the plurality of redefined rules to a plurality of insurance applications in the case base; and

comparing the underwriting decisions in the case base and the underwriting decisions performed by applying the plurality of redefined rules.

4. The computer implemented process according to claim 1 , where the step of comparing further comprises generating a first matrix.

5. The computer implemented process according to claim 1 , where the step of comparing further comprises generating a first matrix comprising values associated with the correlation between the first underwriting decisions and the second underwriting decisions, and the generating at least one penalty further comprises the steps of:

generating a second matrix based on actuarial information and comprising values associated with corresponding degrees of misclassifications between the first underwriting decisions and the corresponding second underwriting decision; and

performing an element-by-element multiplication of the cells of the second matrix with the cells of the first matrix to generate an aggregate penalty.

6. The computer implemented process according to claim 1 , where the process evaluates medical data included in at least one of the plurality of insurance applications.

7. A computer readable medium storing code for causing a computer device to optimize an insurance application underwriting process based on a plurality of predetermined underwriting rules for an insurance underwriting system, the medium comprising:

code that renders a first underwriting decision for each insurance application based on respectively applying a first process to the parameters of each insurance application of a plurality of insurance applications, the first underwriting decision comprising a first classification of each insurance application, the first underwriting decisions being designated as included in a case base;

code that defines a second process constituted by a plurality of underwriting rules for the insurance underwriting system;

code that renders, by the computer device, a second underwriting decision for each insurance application based on respectively applying the second process to the parameters of the each insurance application, the second underwriting decision comprising a second classification of each insurance application, the second process utilizing a plurality of fuzzy rules utilizing decision thresholds;

code that compares, to perform a comparison of, the first underwriting decisions and the second underwriting decisions to generate a result, the result relating to a degree of matching between the first underwriting decision rendered for an insurance application and the second underwriting decision rendered for the same application;

code that generates at least one penalty based on the comparison of the first underwriting decisions and the second underwriting decisions, the generates at least one penalty including aggregating (1) penalties for misclassifications, between the second underwriting decisions and the first underwriting decisions, with (2) rewards for correct classifications, between the second underwriting decisions and the first underwriting decisions;

wherein penalties for misclassification of rate classes that a second underwriting decision is as compared to a corresponding first underwriting decisions;

where the first process and the second process are a plurality of underwriting rules; and

code that outputs the result of the comparison between the first underwriting decision and the second underwriting decision performed by the process; and

code that redefines the plurality of underwriting rules of the second process based on the generated penalty and the comparison of the first underwriting decisions and the second underwriting decisions performed by applying the plurality of underwriting rules, the redefining comprising of tuning the decision thresholds of the fuzzy rules.

8. The computer readable medium according to claim 7 , where the code that compares further comprises generating a first matrix comprising values associated with the correlation between the first underwriting decisions and the second underwriting decisions rendered by applying the second process.

9. The computer readable medium according to claim 7 , where the code for generating at least one penalty further comprises:

code that generates a second matrix based on actuarial information and comprising values associated with corresponding degrees of misclassifications between the first underwriting decisions and the corresponding second underwriting decisions rendered by applying the second process; and

code that performs an element-by-element multiplication of the cells of the second matrix with the cells of the first matrix to generate an aggregate penalty.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 4, 2017
From: GENWORTH HOLDINGS, INC.
To: PACIFIC LIFE INSURANCE COMPANY
Reel/Frame 043783/0654 →
MERGER Recorded May 23, 2013
From: GENWORTH FINANCIAL, INC.
To: GENWORTH HOLDINGS, INC.
Reel/Frame 030485/0945 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 22, 2004
From: GE FINANCIAL ASSURANCE HOLDINGS, INC.
To: GENWORTH FINANCIAL, INC.
Reel/Frame 015165/0513 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 18, 2002
From: BONISSONE, PIERO PATRONE; MESSMER, RICHARD PAUL; PATTERSON, ANGELA NEFF; YANG, DAN; PAVESE, MARC; SUBBU, RAJESH VENKAT; AGGOUR, KAREEM SHERIF
To: GE FINANCIAL ASSURANCE HOLDINGS, INC.
Reel/Frame 013016/0179 →