Express Scripts Strategic Development, Inc.

Morgan J Finley

Garret L Anderson

Camille Patel

John C Petch

Billeigh Barackman

Stephen J Bergner

Christopher Greer

Ryann Ewing

Sai Pranay Malipatil

Michael Nassar

Siju Vattakunnumpurath Eugin

Daniel Owens

Nihitha Davis

Heather Durosko

A machine learning server for trained prediction of unenrollment in claim processing systems is provided. The machine learning server includes a processor configured to construct a layered neural network configured to be trained by claim features. The processor is also configured to receive historic claims. The processor is also configured to extract a set of historic claim features from the historic claim data of the historic claims. The processor is further configured to obtain the claim enrollment status associated with the sets of historic claim features. The processor is also configured to train the layered neural network with the set of historic claim features and the claim enrollment status to predict the claim enrollment status of a pending claim. The processor is further configured to receive a pending claim and to apply the pending claim features to the trained layered neural network to obtain a predicted claim enrollment status.

Machine learning systems for predicting unenrollment in claims processing

A method for predicting errors in prescription claim data is performed by a claim analysis device. The method includes extracting historical claim features from successfully processed historical claims received from the data warehouse system. The method includes extracting pending claim features from a pending claim. The method includes applying a binarization process on the extracted historical claim features to obtain a binarized training feature set. The method includes applying the binarization process on the extracted pending claim features to obtain a binarized pending feature set. The method includes calculating an aggregate distance between the binarized pending feature set and the binarized training feature set. The method includes identifying the historical claim associated with the least aggregate distance as a predictive historical claim. The method includes transmitting an alert upon determining that a billing attribute of the predictive historical claim fails to match a corresponding billing attribute of the pending claim.

Machine learning systems for error detection in data processing systems and related methods

A method includes receiving at least one data object and identifying, for a first aspect of the at least one data object, text strings of free form textual information having a first text string type. The method also includes generating updated free form textual information by removing the at least one text string having the first text string type. The method also includes generating one or more feature vectors based on the updated free form textual information using at least one of a unigram, a bigram, and a trigram. The method also includes using an artificial intelligence engine that uses at least one machine learning model configured to provide, using the one or more feature vectors, an output that includes at least one prediction indicating at least one resource domain and a weight value indicating a probability that the at least one resource domain corresponds to the free form textual information.

System and method for natural language processing

Growing community of inventors

St Louis Park, MN, United States of America

Morgan J Finley