Artificial intelligence automation for enrollment

Patent metadata

Field	Value
Publication number	US-11537913-B2
Application number	US-202016816996-A
Country	US
Kind code	B2
Filing date	Mar 12, 2020
Priority date	Mar 12, 2020
Publication date	Dec 27, 2022
Grant date	Dec 27, 2022

Abstract

A system and method are disclosed for detecting anomalies in input data. The input data can include enrollment information for members of a health insurance product offered by an insurance carrier. A system is configured to process one or more parameters associated with the input data to detect whether an anomaly is indicated in the input data. At least one artificial intelligence algorithm is configured to compare a parameter in the one or more parameters to a range defined by a moving average. When an anomaly is detected, the processing can be paused and manual intervention can be facilitated. When no anomaly is detected, a database is updated based on the input data. In an embodiment, the input data is an ANSI 834 format file that includes a plurality of membership records, and the one or more parameters are derived from statistics related to the membership records.

First claim

What is claimed is:

1. A system for detecting anomalies, the system comprising:
a memory storing input data in a first format; and
one or more processors associated with a health insurance carrier and configured to:
process one or more parameters associated with the input data to detect whether an anomaly is indicated in the input data, wherein at least one artificial intelligence algorithm is configured to compare a parameter in the one or more parameters to a range defined by a moving average; and
pause the processing in response to detecting the anomaly, or
update, responsive to not detecting the anomaly, a database based on the input data,

wherein the input data is one of:
an ANSI 834 format file; or
an ANSI 837 format file; and

wherein the one or more processors are further configured to:
receive the input data from a plan sponsor via a network;
generate a data structure associated with the input data that includes at least one of the one or more parameters processed by the at least one artificial intelligence algorithm;
convert the input data to a second format based, at least in part, on one or more validation rules; and
store the input data in the second format in the data structure.

2. The system of claim 1 , wherein the one or more parameters are derived from a plurality of membership records included in the input data.

3. The system of claim 1 , wherein the at least one artificial intelligence algorithm is included in an artificial intelligence service, wherein at least one processor in the one or more processors is configured to call the artificial intelligence service via an application programming interface (API), and wherein the call includes the one or more parameters.

4. The system of claim 1 , wherein the moving average is calculated, based on a sliding window, as a sum of values for the parameter received during the sliding window from a plurality of input data divided by a number of values included in the sum, and wherein the range is calculated by multiplying the moving average by a minimum coefficient and a maximum coefficient.

5. The system of claim 1 , wherein an upper limit and a lower limit of the range are learned parameters based on training the artificial intelligence algorithm.

6. The system of claim 1 , wherein the one or more parameters include at least one of: a number of records in the input data, a number of new enrollments in the input data, a number of existing enrollments in the input data, or a number of errors in demographics data included in the input data.

7. The system of claim 1 , wherein the one or more processors are further configured to implement a user interface that generates a notification transmitted to a client device that indicates the anomaly was detected in the input data.

8. The system of claim 1 , wherein the one or more parameters include a plurality of parameters, wherein each parameter in the plurality of parameters is processed by a different artificial intelligence algorithm in a plurality of artificial intelligence algorithms, and wherein each artificial intelligence algorithm utilizes different coefficients to calculate a range to compare with a corresponding parameter in the plurality of parameters.

9. A method, comprising:
receiving input data in a first format;
processing, via at least one artificial intelligence algorithm, one or more parameters derived from the input data, wherein each artificial intelligence algorithm in the at least one artificial intelligence algorithm is configured to compare a value for a parameter to a range calculated based on a moving average; and
detecting an anomaly in the input data based on an output of the at least one artificial intelligence algorithm,
wherein the input data is received from a plan sponsor via a network and is one of:
an ANSI 834 format file; or
an ANSI 837 format file; and

wherein the method further comprises:
generating a data structure associated with the input data that includes at least one of the one or more parameters processed by the at least one artificial intelligence algorithm,
converting the input data to a second format based, at least in part, on one or more validation rules; and
storing the input data in the second format in the data structure.

10. The method of claim 9 , wherein the moving average is calculated, based on a sliding window, as a sum of values for the parameter received during the sliding window from a plurality of input data divided by a number of values included in the sum, and wherein the range is calculated by multiplying the moving average by a minimum coefficient and a maximum coefficient.

11. The method of claim 9 , wherein an upper limit and a lower limit of the range are learned parameters based on training the artificial intelligence algorithm.

12. The method of claim 9 , the method further comprising:
generating a notification transmitted to a client device that indicates the anomaly was detected in the input data.

13. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to detect anomalies by performing steps comprising:
receiving input data in a first format;
processing, via at least one artificial intelligence algorithm, one or more parameters derived from the input data, wherein each artificial intelligence algorithm in the at least one artificial intelligence algorithm is configured to compare a value for a parameter to a range calculated based on a moving average; and
detecting an anomaly in the input data based on an output of the at least one artificial intelligence algorithm,
wherein the input data is received from a plan sponsor via a network and is one of:
an ANSI 834 format file; or
an ANSI 837 format file; and

wherein the method further comprises:
generating a data structure associated with the input data that includes at least one of the one or more parameters processed by the at least one artificial intelligence algorithm,
converting the input data to a second format based, at least in part, on one or more validation rules; and
storing the input data in the second format in the data structure.

14. The computer-readable medium of claim 13 , wherein the moving average is calculated, based on a sliding window, as a sum of values for the parameter received during the sliding window from a plurality of input data divided by a number of values included in the sum, and wherein the range is calculated by multiplying the moving average by a minimum coefficient and a maximum coefficient.

Assignees

• Aetna Inc

Inventors

• Johnson Rajesh
• Joglekar Mrinal M
• Potter-Watkins Jamie L

Classifications

• G06Q40/08

  Insurance · CPC title

• G06F16/258

  Data format conversion from or to a database · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06F16/2379

  Updates performed during online database operations; commit processing · CPC title

• G06N5/04 Primary

  Inference or reasoning models · CPC title