US 11,961,598 B1
Machine learning systems for error detection in data processing systems and related methods
Morgan J. Finley, St. Louis Park, MN (US); Garret L. Anderson, Arden Hills, MN (US); Camille Patel, New York, NY (US); Michael Nassar, Chicago, IL (US); Siju Vattakunnumpurath Eugin, Parsippany, NJ (US); and Daniel Owens, Pompton Plains, NJ (US)
Assigned to Express Scripts Strategic Development, Inc., St. Louis, MO (US)
Filed by Express Scripts Strategic Development, Inc., St. Louis, MO (US)
Filed on Jun. 26, 2020, as Appl. No. 16/913,797.
Int. Cl. G16H 20/10 (2018.01); G06F 16/28 (2019.01); G06N 20/00 (2019.01); G16H 40/20 (2018.01)
CPC G16H 20/10 (2018.01) [G06F 16/283 (2019.01); G06N 20/00 (2019.01); G16H 40/20 (2018.01)] 20 Claims
OG exemplary drawing
 
8. A method for use in dispensing one or more prescription drugs, said method performed by a claim analysis device and a pharmacy fulfillment device in communication with a data warehouse system and a claim processing system, the data warehouse system including historical claims including a set of historical claim features and a historical claim status, the claim processing system including pending claims including a set of pending claim features, the pharmacy fulfillment device configured to dispense a prescription drug, the claim analysis device including a processor and a memory, said method comprising:
extracting, by the claim analysis device, subsets of historical claim features from a subset of historical claims received from the data warehouse system, wherein each of the subset of historical claims has a corresponding historical claim status indicating that the claim has been successfully processed;
extracting, by the claim analysis device, a subset of pending claim features from a pending claim received from the claim processing system;
applying, by the claim analysis device, a binarization process on the extracted subsets of historical claim features to obtain a binarized training feature set for the subset of historical claims;
applying, by the claim analysis device, the binarization process on the extracted subset of pending claim features to obtain a binarized pending feature set for the pending claim;
calculating, by the claim analysis device, an aggregate distance between the binarized pending feature set and the binarized training feature set for the subset of historical claims;
identifying, by the claim analysis device, from the subset of historical claims, a historical claim associated with the least aggregate distance as a predictive historical claim;
determining, by the claim analysis device, that a billing attribute of the predictive historical claim matches a corresponding billing attribute of the pending claim;
validating, by the claim analysis device, the pending claim and transmitting, to the pharmacy fulfillment device, an instruction for dispensing the prescription drug in accordance with the pending claim; and
dispensing, by the pharmacy fulfillment device, the prescription drug based on the instruction from the claim analysis device.