Data mining based underwriting profitability analysis

• US 5,970,464 A
• Filed: 09/10/1997
• Issued: 10/19/1999
• Est. Priority Date: 09/10/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Tipu, Fateh Ali, Wang, Hsueh-ju, Rosen, Barry Kumin, Apte, Chidanand Vinayak, White, Brian Frederick, Pednault, Edwin Peter Dawson, Grossman, Edna
• Primary Examiner(s): Tkacs, Stephen R.
• Assistant Examiner(s): Morgan, George D.
• Application Number:
  US08/926,804
• Time in Patent Office: 769 Days
• Field of Search: 705/4, 705/7, 705/10, 705/35, 705/36, 705/38, 705/2, 706/934, 706/12, 706/21, 706/60, 706/47, 707/104
• US Class Current: 705/4
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

A computer implemented method of underwriting profitability analysis delivers the analytic process to a wide cross section of insurance decision makers. The underwriting profitability analysis system leverages an existing investment in databases and improves underwriting business processes. Data mining techniques are applied to historical policy and claims to extract rules that describe policy holders with homogeneous claim frequency and severity characteristics. These rule sets are used to classify policy holders into distinct risk groups, each with its own set of characteristics, including pure premium. Breaking up a book of business into segments allows identification of sub-populations of policy holders that distinctly deviate from the expected normal pure premium. This identification allow the insurance business analysts to interactively adjust eligibility criteria and examine altered characteristics of the covered segments until satisfactory. The system is implemented on a client server using network centric language technology.

Claims

1. A computer implemented method of underwriting profitability analysis comprising the steps of:

accessing existing databases of historical policy and claims;

initiating data mining techniques to the accessed data to extract rules that describe policy holders with homogeneous claim frequency and claim severity characteristics;

using the extracted rules to classify policy holders into distinct risk groups, each with its own set of characteristics, including pure premium; and

identifying sub-populations of policy holders that distinctly deviate from the expected normal pure premium to allow the insurance business analysts to interactively adjust the eligibility criteria and examine altered characteristics of the covered risk groups.

2. The computer implemented method of underwriting profitability analysis recited in claim 1 wherein the method is implemented on a client server platform that employs network centric language technology, thereby allowing a server based analytic methodology that can be simultaneously accessed across an intranet or Internet by several decision makers, from diverse function areas in a firm.

3. The computer implemented method of underwriting profitability analysis recited in claim 2 wherein the method is implemented using an Internet enabled software platform wherein the application interface is controllable by a user running a light weight Internet client and wherein the data intensive and computer intensive work is performed on a server on a network connected to the Internet.

4. A computer implemented method of underwriting profitability analysis comprising the steps of:

accessing existing databases of historical polic and claims;

initiating data mining techniques to the accessed data to extract rules that describe policy holders with homogeneous claim frequency and claim severity characteristics;

using the extracted rules to classify policy holders into distinct risk groups, each with its own set of characteristics, including pure premium; and

identifying sub-populations of policy holders that distinctly deviate from the expected normal pure premium to allow the insurance business analysts to interactively adjust the eligibility criteria and examine altered characteristics of the covered risk groups,

wherein the step of accessing existing databases comprises the steps of;

displaying databases for selection by a user;

responding to a user selection of a database by requesting data from data storage and computing univariate statistics;

updating and parsing meta-data and returning updated and parsed meta-data in response to the data request; and

generating and displaying a data summary of the returned updated and parsed meta-data.

5. The computer implemented method of underwriting profitability analysis recited in claim 4 wherein the step of initiating data mining techniques comprises the steps of:

responding to a user selection of a data set for a data mining operation by mining data in the selected data set to produce policy pure premium prediction rules and storing the rules; and

displaying a data summary of rule sets from the data mining operation.

6. The computer implemented method of underwriting profitability analysis recited in claim 5 wherein the step of using the extracted rules to classify policy holders into distinct risk groups comprises the steps of:

responding to a user selection of a rule set by displaying summary statistics of the selected rule set; and

editing the rule set in response to editing commands from the user.

7. The computer implemented method of underwriting profitability analysis recited in claim 6 wherein the step of identifying sub-populations of policy holders comprises the steps of:

receiving user input specified data set, rule set and product eligibility criteria;

segmenting the data set by eligibility criteria and rule set; and

displaying a segmentation table of the segmented data set.