System for synergistic data processing

• US 20080077451A1
• Filed: 09/24/2007
• Published: 03/27/2008
• Est. Priority Date: 09/22/2006
• Status: Abandoned Application

• 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: Anthony, John J., Borden, Richard M.
• Application Number:
  US11/903,927
• Publication Number:
  US 20080077451A1
• Time in Patent Office: Days
• US Class Current: 705/4
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

A data analysis system that includes an information mining engine for extracting structured data from unstructured data, a data store for storing the extracted structured data, data received from third party data sources, and data received from sensors monitoring insured property is described. The system also includes a business logic processor that synergistically analyzes the structured data extracted by the text mining engine, the data received from the sensor, and the data received from the third party data source to make an insurance evaluation.

Claims

1. A data analysis system comprising:

an information mining engine for extracting structured data from unstructured information;

a data store for storing the structured data output by the information mining engine, the data store further configured to receive data from a sensor monitoring an insured property and data from a third party data source; and

a business logic processor for collectively analyzing the structured data extracted by the information mining engine, the data received from the sensor, and the data received from the third party data source to make an insurance evaluation.

2. The data analysis system of claim 1, comprising a relationship engine configured to identify linkages between data fields stored in the data store.

3. The data analysis system of claim 2, wherein the relationship engine is configured to identify a linkage between a data field stored in the data store and a third party data source from which data is available to populate the data field.

4. The data analysis system of claim 2, wherein the relationship engine is configured to identify a linkage between a data field stored in the data store and the sensor monitoring the insured property in order to obtain data to populate the data field.

5. The data analysis system of claim 1, wherein the business logic processor comprises a predictive model for detecting fraud in an insurance claim based on a combination of the structured data extracted by the information mining engine, the data obtained from the sensor, and the data collected from the third party data source.

6. The data analysis system of claim 1, wherein the business logic processor comprises a predictive model for detecting fraud in an application for insurance based on a combination of the structured data extracted by the information mining engine, the data obtained from the sensor, and the data collected from the third party data source.

7. The data analysis system of claim 1, wherein the business logic processor comprises a predictive model for evaluating a loss associated with an insurance claim based on a combination of the structured data extracted by the information mining engine, the data obtained from the sensor, and the data collected from the third party data source.

8. The data analysis system of claim 1, wherein the business logic processor comprises a predictive model for underwriting an application for insurance based on a combination of the structured data extracted by the information mining engine, the data obtained from the sensor, and the data collected from the third party data source.

9. The data analysis system of claim 1, wherein the business logic processor comprises a predictive model that, in relation to a condition identified by the information mining engine, evaluates the import of collected sensor data based on data retrieved from a third party.

10. The data analysis system of claim 1, wherein the information mining engine comprises an image mining engine for extracting structured data from images or video.

11. A method of making an insurance evaluation comprising:

receiving data from an information mining engine, a third party data source, and a telematics sensor;

collectively processing the received data by a business logic processor including a predictive model; and

determining one of a likelihood of insurance fraud, a premium price, an underwriting rating, an estimated ultimate severity, and a likelihood of subrogation using the predictive model based on the combination of the data received from the information mining engine, the third party data source, and the telematics sensor.

12. The method of claim 11, comprising altering a step in an insurance work flow based on the determination.

13. The method of claim 11, comprising analyzing data received from the telematics sensor to verify data received from the information mining engine.

14. The method of claim 11, comprising analyzing data received from the telematics sensor and the third party data source to verify data received from the information mining engine.

15. The method of claim 11, comprising analyzing data received from the third party data source to verify data received from the information mining engine.

16. The method of claim 11, comprising analyzing data received from the third party data source and the third party data source to verify data received from the information mining engine.

17. The method of claim 11, wherein receiving data from the third party data source comprises receiving data from the third party data source based on the data received from the telematics sensor.

18. The method of claim 17, wherein the data from the third party data source is used to interpret the data received from the telematics sensor in relation to a condition identified by the information mining engine.

19. The method of claim 11, wherein receiving data from the third party data source comprises:

identifying at least one data field utilized by the predictive model for which data is not currently stored in a data store;

identifying the third party data source from which the data to populate the data field is available; and

querying the identified third party data source using the data received from the telematics sensor to obtain the data from the third party data source.

20. The method of claim 11, wherein receiving data from the third party data source comprises:

identifying at least one data field utilized by the predictive model for which data is not currently stored in a data store;

identifying the third party data source from which the data to populate the data field is available; and

querying the identified third party data source using the data received from the telematics sensor and the data received from the information mining engine to obtain the data from the third party data source.

21. The method of claim 11, comprising updating the predictive model based on the received data.

22. The method of claim 11, wherein the information mining engine comprises a text mining engine for extracting structured data from unstructured text.

23. A computer readable medium having computer-executable instructions for making insurance evaluations stored thereon, said computer-executable instructions, upon execution by a computer apparatus, cause the computer apparatus to perform:

receiving data from a telematics sensor, an information mining engine, and a third party data source, and;

processing the received data by a business logic processor including a predictive model; and

determining one of a likelihood of insurance fraud, a premium price, an underwriting rating, an estimated ultimate severity, and a likelihood of subrogation using the predictive model based on the combination of the data received from the information mining engine, the third party data source, and the telematics sensor.

24. The computer readable medium of claim 23, wherein obtaining data from the third party data source comprises:

identifying at least one data field utilized by the predictive model for which data is not currently stored in a data store;

identifying the third party data source from which the data to populate the data field is available; and

querying the identified third party data source using the data received from the telematics sensor and the data received from the information mining engine to obtain the data from the third party data source.

25. The computer readable medium of claim 23, wherein receiving data from the third party data source comprises receiving data from the third party data source based on the data received from the telematics sensor, and the data received from the third party data source is used to interpret the data received from the telematics sensor in relation to a condition identified by the information mining engine.