US 11,669,907 B1
Methods and apparatus to process insurance claims using cloud computing
Mercy Behrens, Bloomington, IL (US); Donald Joseph Yuhas, Bloomington, IL (US); and Jason Beckman, Bloomington, IL (US)
Assigned to State Farm Mutual Automobile Insurance Company, Bloomington, IL (US)
Filed by STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY, Bloomington, IL (US)
Filed on Jun. 27, 2019, as Appl. No. 16/455,533.
Int. Cl. G06Q 40/08 (2012.01); G06N 20/00 (2019.01); H04L 67/10 (2022.01); G06F 40/205 (2020.01)
CPC G06Q 40/08 (2013.01) [G06F 40/205 (2020.01); G06N 20/00 (2019.01); H04L 67/10 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system to process an insurance claim using cloud computing, the system comprising:
an assistance resource of a cloud computing platform, wherein the assistance resource is configured to:
detect an event related to an insurance claim, the insurance claim being characterized by a particular claim type; and
determine, based on detection of the event, a status of the insurance claim, wherein the status is indicative of a number of incomplete steps in processing the insurance claim;
an artificial intelligence resource of the cloud computing platform, wherein the artificial intelligence resource is configured to:
select, based on the claim type and from a plurality of machine learning models associated with respective claim types, a machine learning model configured to identify a next action to be taken in processing the insurance claim, wherein:
the selected machine learning model comprises a model trained to analyze data of insurance claims characterized by the claim type, and
the model is trained on training data comprising sequences of processing actions taken in previously-processed insurance claims of the claim type;
determine, using the selected machine learning model, a processing action to be taken for the insurance claim based at least in part on the status of the insurance claim, the processing action being associated with at least one of the incomplete steps;
predict, based on detection of the event and for each skill module of a plurality of skill modules associated with a backend system, a confidence level indicative of an ability to perform the processing action without receiving manual input;
select, automatically and without receiving manual input, a skill module of the plurality of skill modules based on the predicted confidence levels; and
provide an executable command, corresponding to the processing action, to the skill module, the command causing the skill module to:
perform the processing action; and
update the status of the insurance claim based on completion of the processing action.