System and method for implementing computer modeling techniques

• US 10,332,210 B1
• Filed: 11/06/2014
• Issued: 06/25/2019
• Est. Priority Date: 11/06/2013
• Status: Active Grant

• Current Assignee: Nationwide Mutual Insurance Company
• Original Assignee: Nationwide Mutual Insurance Company
• Inventors: Kemp, John, Herath, Spencer, Winningham, Thomas Michael, Terry, Shannon, Nicholas, Christopher P., Choi, Kapwon, Ogorek, Benjamin A.
• Primary Examiner(s): Wong, Eric T
• Application Number:
  US14/534,437
• Time in Patent Office: 1,692 Days
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

Improved computer modeling techniques, and uses thereof, are described herein. A set of unstructured textual data is received. Certain textual data is removed from the set of unstructured data to form an initial vocabulary set of textual data. One or more bigrams are added to the vocabulary set of textual data to form a final vocabulary set of textual data. The final vocabulary set of textual data is divided into a plurality of subsets of textual data based on type. A model is trained using each of the plurality of subsets of textual data to form a plurality of trained models, each corresponding to one of the types.

Claims

1. A computer implemented method comprising:

receiving at a computer processor a set of unstructured textual property claims data corresponding to a plurality of property loss insurance claims, wherein the set of unstructured textual property claims data comprises data describing claims associated with subrogation and claims not associated with subrogation, wherein a number of the claims associated with subrogation is the same as a number of the claims not associated with subrogation;

removing, using the computer processor, at least certain textual data from the set of unstructured textual property claims data to form an initial vocabulary set of textual data;

adding, using the computer processor, one or more bigrams to the vocabulary set of textual data to form a final vocabulary set of textual data;

identifying, using the computer processor, a property loss type for each of the plurality of property loss insurance claims;

dividing, using the computer processor, the final vocabulary set of textual data into a plurality of subsets of textual data based on the property loss type;

training, using the computer processor, a model using each of the plurality of subsets of textual data to form a plurality of trained models, each corresponding to one of the property loss types receiving at the computer processor data describing a property insurance loss claim;

determining, using the computer processor, whether the property insurance loss claim is a qualifying claim;

identifying, using the computer processor, a property loss type associated with the property insurance loss claim;

selecting, using the computer processor, one of the plurality of models based on the identified property loss type;

processing, using the computer processor, the data describing the property insurance loss claim using the selected model; and

determining, using the computer processor, whether to analyze the property insurance loss claim for subrogation based on the processing.

2. The computer implemented method of claim 1 wherein the certain textual data removed from the set of unstructured textual data comprises one or more of numbers, punctuation, stop words, and context-specific words that occur within the set of unstructured textual property claims data with a predetermined level of frequency.

3. The computer implemented method of claim 1 wherein the property loss type comprises one of water, fire and damage by vehicle.

4. The computer implemented method of claim 1 wherein the determining comprises determining whether the property insurance loss claim was initiated within the past 7 to 14 days.

5. A non-transitory computer readable storage medium having stored thereon computer executable instructions that, when executed on a computer, configure the computer to perform a method comprising:

receiving a set of unstructured textual property claims data corresponding to a plurality of property loss insurance claims, wherein the set of unstructured textual property claims data comprises data describing claims associated with subrogation and claims not associated with subrogation, wherein a number of the claims associated with subrogation is the same as a number of the claims not associated with subrogation;

removing at least certain textual data from the set of unstructured textual property claims data to form an initial vocabulary set of textual data;

adding one or more bigrams to the vocabulary set of textual data to form a final vocabulary set of textual data;

identifying a property loss type for each of the plurality of property loss insurance claims;

dividing the final vocabulary set of textual data into a plurality of subsets of textual data based on the property loss type;

training a model using each of the plurality of subsets of textual data to form a plurality of trained models, each corresponding to one of the property loss types;

receiving data describing a property insurance loss claim;

determining whether the property insurance loss claim is a qualifying claim;

identifying a property loss type associated with the property insurance loss claim;

selecting one of the plurality of models based on the identified property loss type;

processing the data describing the property insurance loss claim using the selected model; and

determining whether to analyze the property insurance loss claim for subrogation based on the processing.

6. The non-transitory computer readable storage medium of claim 5 wherein the certain textual data removed from the set of unstructured textual data comprises one or more of numbers, punctuation, stop words, and context-specific words that occur within the set of unstructured textual property claims data with a predetermined level of frequency.

7. The non-transitory computer readable storage medium of claim 5 wherein the property loss type comprises one of water, fire and damage by vehicle.

8. The non-transitory computer readable storage medium of claim 5 wherein the determining comprises determining whether the property insurance loss claim was initiated within the past 7 to 14 days.

9. A system comprising:

memory operable to store at least one program; and

at least one processor communicatively coupled to the memory, in which the at least one program, when executed by the at least one processor, causes the at least one processor to perform a method comprising;

receiving a set of unstructured textual property claims data corresponding to a plurality of property loss insurance claims, wherein the set of unstructured textual property claims data comprises data describing claims associated with subrogation and claims not associated with subrogation, wherein a number of the claims associated with subrogation is the same as a number of the claims not associated with subrogation;

removing at least certain textual data from the set of unstructured textual property claims data to form an initial vocabulary set of textual data;

adding one or more bigrams to the vocabulary set of textual data to form a final vocabulary set of textual data;

identifying a property loss type for each of the plurality of property loss insurance claims;

dividing the final vocabulary set of textual data into a plurality of subsets of textual data based on the property loss type; and

training a model using each of the plurality of subsets of textual data to form a plurality of trained models, each corresponding to one of the property loss types;

receiving data describing a property insurance loss claim;

determining whether the property insurance loss claim is a qualifying claim;

identifying a property loss type associated with the property insurance loss claim;

selecting one of the plurality of models based on the identified property loss type;

processing the data describing the property insurance loss claim using the selected model; and

determining whether to analyze the property insurance loss claim for subrogation based on the processing.

10. The system of claim 9 wherein the certain textual data removed from the set of unstructured textual data comprises one or more of numbers, punctuation, stop words, and context-specific words that occur within the set of unstructured textual property claims data with a predetermined level of frequency.

11. The system of claim 9 wherein the property loss type comprises one of water, fire and damage by vehicle.

12. The system of claim 9 wherein the determining comprises determining whether the property insurance loss claim was initiated within the past 7 to 14 days.