公开(公告)号：US11954602B1

公开(公告)日：2024-04-09

申请号：US16792635

申请日：2020-02-17

Applicant: Optum, Inc.

Inventor： Daniel J. Mulcahy ,  Subhash Seelam ,  Damian Kelly ,  Vijay S. Nori ,  Adam Russell

IPC: G06N5/02 ,  G06F40/284 ,  G06N20/00

CPC classification number: G06N5/02 ,  G06F40/284 ,  G06N20/00

Abstract: There is a need for more effective and efficient predictive data analysis. This need can be addressed by, for example, solutions for performing/executing hybrid input predictive data analysis. In one example, a method includes identifying a vocabulary data object associated with one or more prediction input text data objects; determining, based at least in part on the vocabulary data object, a per-input-entity tokenized representation for the one or more prediction input text data objects; determining, based at least in part on the per-input-entity tokenized representation and one or more prediction input structured data objects and using a hybrid-input predictive model, a prediction score; determining, based at least in part on one or more threshold determination configuration criteria, a predictive threshold for the hybrid-input predictive model; generating, based at least in part on the prediction score and the predictive threshold, a predictive output; and performing one or more prediction-based actions based at least in part on the predictive output.

公开(公告)号：US11537818B2

公开(公告)日：2022-12-27

申请号：US16746232

申请日：2020-01-17

Applicant: Optum, Inc.

Inventor： Vijay S. Nori ,  Christopher A. Hane ,  Paul A. Bleicher

IPC: G06K9/00 ,  G06K9/62 ,  G06F17/16 ,  G06N20/00

Abstract: Various embodiments of the disclosure provide apparatuses, systems, and computer program products for predictive data labelling using a dual-model system. Embodiments provide various advantages in accuracy of predicted labels, for example in various contexts such as medical data analysis for difficult to diagnose diseases. An example provided apparatus is configured to generate a positive, neutral, and negative candidate identifier sets and corresponding positive, neutral, and negative candidate index sets based in part on applying a candidate selection rule set to a candidate data set; train a candidate label probabilistic model based at least in part on a candidate label training subset associated with the candidate data set associated with the positive and negative candidate identifiers; generate a candidate positive-label probability set using at least the candidate label probabilistic model; train a historical record prediction model to predict the candidate positive-label probability set; and utilize the historical record prediction model.

公开(公告)号：US20210294981A1

公开(公告)日：2021-09-23

申请号：US17303678

申请日：2021-06-04

Applicant: OPTUM, INC.

Inventor： Christopher A. Hane ,  Vijay S. Nori ,  Louis J. Rumanes

IPC: G06F40/30

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy. This need can be addressed by, for example, by identifying an indexed representation of a natural language object; obtaining a vocabulary domain associated with one or more first phrases; determining an individual frequency for each first phrase based on a count of occurrences of the first phrase in the indexed representation; identifying one or more dominant phrases of the first phrases; for each dominant phrase, identifying any dependent phrases for the first dominant phrase; determining a semantically-adjusted frequency for each dominant phrase based on the individual frequency for the dominant phrase and each individual frequency for any dependent phrase for the dominant phrase; generating a structured representation of the natural language object based on each semantically-adjusted frequency associated with a dominant phrase; and providing the structured representation for the predictive analysis.

公开(公告)号：US20210240720A1

公开(公告)日：2021-08-05

申请号：US16777784

申请日：2020-01-30

Applicant: Optum, Inc.

Inventor： Jon Kevin Muse ,  Gregory J. Boss ,  Vijay S. Nori ,  Martijn P. Van Overbeek

IPC: G06F16/2457 ,  G06F16/28 ,  G06F16/29 ,  G06F16/2455

Abstract: Various methods and systems for selectively and securely sharing user data to a facility in order to accommodate the specific needs of the user. The methods further correspond to receiving, from a computing entity, geographic location information corresponding to the geographic location of the computing entity which is associated with the user and transmitting a notification to the computing entity of a facility in proximity to the geographic location of the computing entity. The methods further include receiving, from the facility, a request for user data associated with the user of the computing entity that is applicable to the facility, generating a proposed user dataset in response to the request that satisfies the facility-specific user data parameters and transmitting the proposed user dataset that meets the facility-specific user data parameters for sharing with the facility when a relevance score exceeds a relevance threshold value and the sharing eligibility is approved.

公开(公告)号：US20200233928A1

公开(公告)日：2020-07-23

申请号：US16253886

申请日：2019-01-22

Applicant: Optum, Inc.

Inventor： Christopher A. Hane ,  Vijay S. Nori ,  Louis J. Rumanes

IPC: G06F17/27

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy. This need can be addressed by, for example, by identifying an indexed representation of a natural language object; obtaining a vocabulary domain associated with one or more first phrases; determining an individual frequency for each first phrase based on a count of occurrences of the first phrase in the indexed representation; identifying one or more dominant phrases of the first phrases; for each dominant phrase, identifying any dependent phrases for the first dominant phrase; determining a semantically-adjusted frequency for each dominant phrase based on the individual frequency for the dominant phrase and each individual frequency for any dependent phrase for the dominant phrase; generating a structured representation of the natural language object based on each semantically-adjusted frequency associated with a dominant phrase; and providing the structured representation for the predictive analysis.

公开(公告)号：US11699042B2

公开(公告)日：2023-07-11

申请号：US17303680

申请日：2021-06-04

Applicant: OPTUM, INC.

Inventor： Christopher A. Hane ,  Vijay S. Nori ,  Louis J. Rumanes

IPC: G06F40/30

CPC classification number: G06F40/30

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy. This need can be addressed by, for example, by identifying an indexed representation of a natural language object; obtaining a vocabulary domain associated with one or more first phrases; determining an individual frequency for each first phrase based on a count of occurrences of the first phrase in the indexed representation; identifying one or more dominant phrases of the first phrases; for each dominant phrase, identifying any dependent phrases for the first dominant phrase; determining a semantically-adjusted frequency for each dominant phrase based on the individual frequency for the dominant phrase and each individual frequency for any dependent phrase for the dominant phrase; generating a structured representation of the natural language object based on each semantically-adjusted frequency associated with a dominant phrase; and providing the structured representation for the predictive analysis.

公开(公告)号：US11699041B2

公开(公告)日：2023-07-11

申请号：US17303679

申请日：2021-06-04

Applicant: OPTUM, INC.

Inventor： Christopher A. Hane ,  Vijay S. Nori ,  Louis J. Rumanes

IPC: G06F40/30

CPC classification number: G06F40/30

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy. This need can be addressed by, for example, by identifying an indexed representation of a natural language object; obtaining a vocabulary domain associated with one or more first phrases; determining an individual frequency for each first phrase based on a count of occurrences of the first phrase in the indexed representation; identifying one or more dominant phrases of the first phrases; for each dominant phrase, identifying any dependent phrases for the first dominant phrase; determining a semantically-adjusted frequency for each dominant phrase based on the individual frequency for the dominant phrase and each individual frequency for any dependent phrase for the dominant phrase; generating a structured representation of the natural language object based on each semantically-adjusted frequency associated with a dominant phrase; and providing the structured representation for the predictive analysis.

公开(公告)号：US11699040B2

公开(公告)日：2023-07-11

申请号：US17303678

申请日：2021-06-04

Applicant: OPTUM, INC.

Inventor： Christopher A. Hane ,  Vijay S. Nori ,  Louis J. Rumanes

IPC: G06F40/30

CPC classification number: G06F40/30

Abstract: There is a need for solutions that perform predictive natural language processing with improved efficiency and/or accuracy. This need can be addressed by, for example, by identifying an indexed representation of a natural language object; obtaining a vocabulary domain associated with one or more first phrases; determining an individual frequency for each first phrase based on a count of occurrences of the first phrase in the indexed representation; identifying one or more dominant phrases of the first phrases; for each dominant phrase, identifying any dependent phrases for the first dominant phrase; determining a semantically-adjusted frequency for each dominant phrase based on the individual frequency for the dominant phrase and each individual frequency for any dependent phrase for the dominant phrase; generating a structured representation of the natural language object based on each semantically-adjusted frequency associated with a dominant phrase; and providing the structured representation for the predictive analysis.

公开(公告)号：US11354319B2

公开(公告)日：2022-06-07

申请号：US16777784

申请日：2020-01-30

Applicant: Optum, Inc.

Inventor： Jon Kevin Muse ,  Gregory J. Boss ,  Vijay S. Nori ,  Martijn P. Van Overbeek

IPC: G06F16/00 ,  G06F16/2457 ,  G06F16/2455 ,  G06F16/29 ,  G06F16/28

Abstract: Various methods and systems for selectively and securely sharing user data to a facility in order to accommodate the specific needs of the user. The methods further correspond to receiving, from a computing entity, geographic location information corresponding to the geographic location of the computing entity which is associated with the user and transmitting a notification to the computing entity of a facility in proximity to the geographic location of the computing entity. The methods further include receiving, from the facility, a request for user data associated with the user of the computing entity that is applicable to the facility, generating a proposed user dataset in response to the request that satisfies the facility-specific user data parameters and transmitting the proposed user dataset that meets the facility-specific user data parameters for sharing with the facility when a relevance score exceeds a relevance threshold value and the sharing eligibility is approved.

公开(公告)号：US12002585B2

公开(公告)日：2024-06-04

申请号：US18062900

申请日：2022-12-07

Applicant: Optum, Inc.

Inventor： Vijay S. Nori ,  Christopher A. Hane ,  Paul A. Bleicher

IPC: G16H50/20 ,  G06F17/16 ,  G06F18/214 ,  G06F18/2415 ,  G06N20/00

CPC classification number: G16H50/20 ,  G06F17/16 ,  G06F18/2155 ,  G06F18/2415 ,  G06N20/00

Abstract: Various embodiments of the disclosure provide apparatuses, systems, and computer program products for predictive data labelling using a dual-model system. Embodiments provide various advantages in accuracy of predicted labels, for example in various contexts such as medical data analysis for difficult to diagnose diseases. An example provided apparatus is configured to generate a positive, neutral, and negative candidate identifier sets and corresponding positive, neutral, and negative candidate index sets based in part on applying a candidate selection rule set to a candidate data set; train a candidate label probabilistic model based at least in part on a candidate label training subset associated with the candidate data set associated with the positive and negative candidate identifiers; generate a candidate positive-label probability set using at least the candidate label probabilistic model; train a historical record prediction model to predict the candidate positive-label probability set; and utilize the historical record prediction model.