| CPC G06Q 40/08 (2013.01) [G06N 5/04 (2013.01); G06N 20/00 (2019.01); G06Q 10/10 (2013.01); G16H 10/60 (2018.01); G16H 40/20 (2018.01); G16H 50/20 (2018.01); G16H 20/10 (2018.01)] | 20 Claims |
|
1. A data processing system comprising:
a processor; and
a machine-readable medium storing executable instructions that, when executed, cause the processor alone or in combination with other processors to perform operations comprising:
obtaining policy coverage information for each of a plurality of insurance policies associated with an insured user using an application programming interface configured to obtain the policy coverage information from one or more insurer portals, the policy coverage information being formatted according to a standard schema for processing policy and claim information;
obtaining insurance claim information associated with a plurality of insurance claims associated with the insured user using the application programming interface configured to obtain the insurance claim information from the one or more insurer portals, the insurance claim information being formatted according to the standard schema for processing policy and claim information;
analyzing the policy coverage information with a policy parsing engine to perform a probabilistic data match on policy coverage descriptions included in the policy coverage information to select standardized descriptions from among a set of standardized insurance information associated with a highest probability of being a match with the policy coverage descriptions and to replace the policy coverage descriptions included in the policy coverage information with the standardized descriptions;
generating standardized policy coverage information from the policy coverage information by converting the policy coverage information from one or more first formats to a standardized format according to a standard schema for processing policy and claim information;
generating standardized insurance claim information from the insurance claim information by converting insurance claim information from one or more second formats to the standardized format according to the standard schema;
grouping the insurance claim information based on events requiring medical treatment by analyzing the standardized insurance claim information using a claims grouping model to obtain event-related claim grouping information, the claims grouping model being trained with first training data formatted according to the standard schema to recognize patterns in insurance claim information indicative of an occurrence of specific types of events requiring medical treatment, the event-related claim grouping information including an indication of one or more events predicted to have occurred based on the insurance claim information and one or more insurance claims of the plurality of insurance claims included in the insurance claim information predicted to be associated with each event of the one or more events;
analyzing each group of claims of the event-related claim grouping information and the standardized policy coverage information using a coverage prediction model to obtain coverage prediction information, the coverage prediction model being trained with second training data formatted according to the standard schema to output a prediction for a group of claims whether one or more insurance policies of the plurality of insurance policies identified in the standardized policy coverage information will cover the group of claims;
providing, via a network connection, the coverage prediction information to a computing device associated with the insured user;
submitting, using the application programming interface, each group of claims for reimbursement to a respective insurer associated with an insurance policy predicted by the coverage prediction model to cover the group of claims;
receiving, for each group of claims, an indication from the respective insurer whether the respective insurer will reimburse costs associated with each group of claims;
determining that reimbursement of a respective group of claims has been rejected based on the indication received from the respective insurer for the respective group of claims; and
fine-tuning the claims grouping model, the coverage prediction model, or both the claims grouping model and the coverage prediction model using a model update unit responsive to determining that reimbursement has been rejected for the respective group of claims.
|