System and method for predictive analysis of crowd sourced data for preemptive loss control

Patent metadata

Field	Value
Publication number	US-10692150-B2
Application number	US-201414243206-A
Country	US
Kind code	B2
Filing date	Apr 2, 2014
Priority date	Apr 2, 2014
Publication date	Jun 23, 2020
Grant date	Jun 23, 2020

Abstract

A crowd sourced based predictive system for detecting and analyzing actuarially significant activity, including high risk activities that may result in damage to an insured property, alerts users of the potential conditions and activities in order to assist with loss control and provide users with the ability to purchase insurance coverages relates to the conditions and activities.

First claim

What is claimed is:

1. A system for intelligently compiling and assessing Internet crowd sourced data to proactively identify and alert policy holders of actuarially-significant events, the system comprising:
a data storage device configured to store crowd sourced data and policy data;
at least one processor coupled to the data storage device;
a communications interface unit coupled to the at least one processor;
a memory coupled to the at least one processor;
and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the at least one processor, the one or more programs including instructions for:
accessing, by the communications interface unit, crowd sourced data on a plurality of crowd sourced data sources on the Internet, said crowd sourced data including at least one of picture data, video data, and text data, wherein some or all of the crowd sourced data includes metadata indicating a location corresponding to the crowd sourced data;
storing the crowd sourced data on the data storage device;
evaluating, by a rules processor configured to apply one or more computerized predictive models, the crowd sourced data to detect one or more actuarially-significant events; and
for each one of the one or more actuarially-significant events:
classifying a severity level for each of the detected actuarially-significant events comprising one of a class 1 severity, class 2 severity, and class 3 severity, wherein events having the class 3 severity have a highest severity;
executing a location routine to perform the steps of:
determining location data for a location associated with one or more crowd sourced data items associated with the detected one of the actuarially-significant events based upon metadata associated with the one or more crowd sourced data items; and
correlating the location data associated with the detected one of the actuarially-significant events with policy data in the data storage device corresponding to an in force coverage policy by comparing the location data with address data associated with in-force coverage policies to determine a policyholder associated with the detected actuarially-significant event;
determining a positive or negative transmit warning flag value associated with the determined policyholder based on factors including a number of detected actuarially-significant events in each of the severity classes and weighting factors applied to at least some of the number of detected actuarially-significant events in each of the severity classes, the weighting factors dependent at least on recency of the detected actuarially-significant event; and
generating, for policyholders having a determined positive transmit warning flag value, for transmittal by the communications interface unit to an address associated with a policyholder computing device based on the actuarially-significant event, an electronic communication including a loss control message corresponding to one or more actuarially-significant events associated with the policyholder;

wherein the loss control based message triggers a claims workflow process related to a first notice of loss;
wherein the claims workflow process comprises initiating a claims adjuster process; and
wherein the claims adjuster process uses the crowd sourced data at least in part to determine a potential claim payment to be made to at least one policyholder.

2. The system of claim 1 , wherein the location based data comprises geotag data.

3. The system of claim 1 , wherein the location based data is correlated to an address of a policyholder.

4. The system of claim 3 , wherein the policyholder address includes a physical address.

5. The system of claim 3 , wherein the policyholder address includes an electronic address.

6. The system of claim 5 , wherein the electronic communication via the system is sent to the electronic address of the policyholder.

7. The system of claim 6 , wherein the electronic address is a mobile phone number and the electronic communication is a text message.

8. The system of claim 1 , wherein issuing an electronic communication via the system further comprises providing a recommendation of one or more coverages related to the crowd sourced data.

9. The system of claim 1 , further comprising offering the listed policyholder one or more additional coverages based on the one or more instances of activities.

10. A computer system for processing crowd sourced data for proactively identifying and alerting policy holders of actuarially significant circumstances, comprising:
an insurance data computer system comprising:
a data storage device configured to store crowd sourced data;
one or more computer processors coupled to the data storage device,
a communications interface unit coupled to the one or more computer processors; and
a computer usable medium, accessible by the one or more computer processors, having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method for identifying actuarially-significant circumstances, said method comprising:
compiling, by the communications interface unit, crowd sourced data on a plurality of real time crowd sourced data sources on the Internet, said crowd sourced data including at least one of images data, text data, and video data, wherein some or all of the crowd sourced data includes metadata indicating a location corresponding to the crowd sourced data;
evaluating, by a rules processor configured to apply one or more computerized predictive models, the crowd sourced data to detect one or more actuarially-significant events; and
for each one of the one or more instances of actuarially significant events:
classifying a severity level for each of the detected actuarially-significant events comprising one of a class 1 severity, class 2 severity, and class 3 severity, wherein events having the class 3 severity have a highest severity;
executing a location routine to perform the steps of:
determining location data for a location associated with one or more crowd sourced data items associated with the detected one of the actuarially-significant events based upon metadata associated with the one or more crowd sourced data items;
correlating the location data associated with the detected one of the actuarially-significant events with policy data in the data storage device corresponding to an in force insurance policy by comparing the determined location data with address data associated with in-force insurance policies to determine a policyholder associated with the detected actuarially-significant event;
determining a positive or negative transmit warning flag value associated with the determined policyholder based on factors including a number of detected actuarially-significant events in each of the severity classes and weighting factors applied to at least some of the number of detected actuarially-significant events in each of the severity classes, the weighting factors dependent at least on recency of the detected actuarially-significant event; and
generating, for policyholders having a determined positive transmit warning flag value, for transmittal by the communications interface unit to an address associated with a policyholder computing device based on the actuarially-significant event, an electronic communication to a policyholder under the in force insurance policy for the actuarially-significant events associated with the policyholder, wherein the electronic communication includes a loss control message; and

an insurance workflow computer in communication with the insurance data system, the insurance workflow computer configured to

Assignees

• Hartford Fire Insurance Comp

Inventors

• Brown Patrick D

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06Q40/08 Primary

  Insurance · CPC title

• G06Q50/01

  Physics · mapped topic