Method and system for classifying documents

• US 8,738,552 B2
• Filed: 08/24/2012
• Issued: 05/27/2014
• Est. Priority Date: 05/31/2006
• Status: Active Grant

• Current Assignee: Hartford Fire Insurance Company (The Hartford Financial Services Group, Inc.)
• Original Assignee: Hartford Fire Insurance Company (The Hartford Financial Services Group, Inc.)
• Inventors: Ellingsworth, Martin E.
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US13/594,432
• Publication Number:
  US 20130110843A1
• Time in Patent Office: 641 Days
• Field of Search: None
• US Class Current: 706/20
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

The invention provides a method and system for classifying insurance files for identification, sorting and efficient collection of subrogation claims. The invention determines whether an insurance claim has merit to warrant claim recovery efforts utilizing software code for partially describing a set of documents having unstructured and structured file data containing terms and phrases having contextual bases, code for transforming the terms and phrases, code for iterating a classification process to determine rules that best classify the set of documents based upon context, code for incorporating the rules into an induction and knowledge representation, thesauri taxonomies and text summarization to classify subrogation claims; code for calculating a base score and a concept vector to identify the selected claims that demonstrate a given probability of subrogation recovery.

Claims

1. A computer system for analyzing data indicative of an insurance claim, comprising:

one or more data storage devices including a database storing synonym groups, wherein each synonym group is associated with at least one concept and each synonym group comprises one or more terms or phrases that are associated with the at least one concept;

one or more computer processors;

a memory in communication with the one or more computer processors and storing program instructions, the one or more computer processors operative with the program instructions to;

receive the data indicative of the insurance claim;

identify one or more terms and phrases within the data indicative of the insurance claim;

compare the identified one or more terms and phrases of the data indicative of the insurance claim to the synonym groups to identify one or more concepts associated with the insurance claim;

generate a score associated with the insurance claim based on said comparison and said identified one or more concepts; and

compare said score with a threshold to determine whether performance of one or more additional tasks relating to the insurance claim is required.

2. The system of claim 1, wherein the one or more computer processors are further operative with the program instructions to, prior to receiving the data indicative of the insurance claim:

receive data indicative of categories of information pertinent to recovery of subrogation claims;

analyze, using a classification process, the data indicative of the categories of information to generate the synonym groups.

3. The system of claim 2, wherein the analyzing using the classification process further comprises identifying patterns of concepts within the data indicative of the categories of information;

wherein the one or more computer processors are further operative to compare the patterns of concepts with the one or more concepts associated with the insurance claim; and

wherein the score associated with the insurance claim is generated further based on said comparison of the patterns of concepts and the one or more concepts associated with the insurance claim.

4. The system of claim 3, wherein each of the patterns of concepts comprises a concept pattern vector, wherein the one or more concepts associated with the insurance claim are formed into an insurance claim concepts vector, and wherein comparing the patterns of concepts with the one or more concepts associated with the insurance claim comprises comparing the concept pattern vector to the insurance claim concepts vector.

5. The system of claim 3, wherein the analyzing using the classification process comprises using an iterative N-Gram analysis to achieve a selected level of accuracy in classifying insurance claims data.

6. The system of claim 3, wherein the analyzing using the classification process comprises creating rules for classifying insurance claims data.

7. The system of claim 1, wherein the one or more computer processors are further operative with the program instructions to identify event messaging data within the data indicative of the insurance claim, and wherein the score associated with the insurance claim is generated further based upon said event messaging data.

8. The system of claim 7, wherein the event messaging data comprises one or more of data relating to a contractual waiver of subrogation rights, data relating to jurisdictional rules, and data relating to subrogation statutes.

9. A computerized method for determining whether an insurance claim merits recovery comprising:

receiving, by one or more computer processors via a communications network, data indicative of the insurance claim;

identifying, by the one or more computer processors, one or more terms and phrases within the data indicative of the insurance claim;

comparing, by the one or more computer processors, the identified one or more terms and phrases of the data indicative of the insurance claim to an index of synonym groups stored within a database, wherein each synonym group is associated with a concept and each synonym group comprises one or more terms or phrases that are associated with the concept;

identifying, by the one or more computer processors, one or more concepts associated with the insurance claim based on said comparison;

generating, by the one or more computer processors, a score associated with the insurance claim based on said comparison and said identified one or more concepts; and

routing data indicative of the score to one or more recipients via said communications network.

10. The method of claim 9, further comprising, prior to receiving the data indicative of the insurance claim:

receiving, by the one or more computer processors, data indicative of categories of information pertinent to recovery of subrogation claims;

analyzing, by the one or more computer processors using a classification process, the data indicative of the categories of information and generating the index of synonym groups.

11. The method of claim 10, wherein the analyzing the data indicative of the categories of information using the classification process further comprises identifying patterns of concepts within the data indicative of the categories of information;

and further comprising;

comparing the patterns of concepts with the one or more concepts associated with the insurance claim;

wherein the score associated with the insurance claim is generated further based on said comparison of the patterns of concepts and the one or more concepts associated with the insurance claim.

12. The method of claim 11, wherein each of the patterns of concepts comprises a concept pattern vector, wherein the one or more concepts associated with the insurance claim are formed into an insurance claim concepts vector, and wherein comparing the patterns of concepts with the one or more concepts associated with the insurance claim comprises comparing the concept pattern vector to the insurance claim concepts vector.

13. The method of claim 11, wherein the analyzing using the classification process comprises using an iterative N-Gram analysis to achieve a selected level of accuracy in classifying insurance claims data.

14. The method of claim 11, wherein the analyzing using the classification process further comprises creating rules for classifying insurance claims data.

15. The method of claim 9, further comprising identifying, by the one or more computer processors, event messaging data within the data indicative of the insurance claim, and wherein the score associated with the insurance claim is generated further based upon said identified event messaging data.

16. The method of claim 15, wherein the event messaging data comprises one or more of data relating to a contractual waiver of subrogation rights, data relating to jurisdictional rules, and data relating to subrogation statutes.

17. A non-transitory computer readable medium having stored therein instructions that, upon execution, cause one or more computer processors to:

receive, by one or more computer processors via a communications network, data indicative of an insurance claim;

identify, by the one or more computer processors, one or more terms and phrases within the data indicative of the insurance claim;

compare, by the one or more computer processors, the identified one or more terms and phrases of the data indicative of the insurance claim to an index of synonym groups stored within a database, wherein each synonym group is associated with a concept and each synonym group comprises one or more terms or phrases that are associated with the concept;

identify, by the one or more computer processors, one or more concepts associated with the insurance claim based on said comparison;

generate, by the one or more computer processors, a score associated with the insurance claim based on said comparison and said identified one or more concepts; and

compare said score with a threshold to determine whether performance of one or more additional tasks relating to the insurance claim is required.

18. The non-transitory computer readable medium of claim 17, wherein the instructions are further configured to, prior to receiving the electronic claim:

receive, by the one or more computer processors, data indicative of categories of information pertinent to recovery of subrogation claims;

analyze, by the one or more computer processors using a classification process, the data indicative of the categories of information and generate the index of synonym groups.

19. The non-transitory computer readable medium of claim 18, wherein the instructions are further configured to:

identify, by the one or more computer processors, concept pattern vectors in the data indicative of the categories of information using the classification process;

compare, by the one or more computer processors, the concept pattern vectors with an insurance claim concepts vector of the one or more concepts associated with the insurance claim;

generate the score associated with the insurance claim further based on said comparison of the concept pattern vectors and the insurance claim concepts vector.

20. The non-transitory computer readable medium of claim 17, wherein the one or more additional tasks comprise generating reports including the generated score and routing the insurance claim to one or more of a litigation department, a collection specialist, a subrogation specialist, and a case manager.