US 12,002,585 B2 · Apparatus, computer program product, and method for predictive data labelling using a dual-prediction model system

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.

Claims20

Dependent claims

A computer-implemented method, the computer-implemented method comprising:

generating, using a first machine learning model, a predicted label probability corresponding to a first label, wherein the predicted label probability indicates a probability that a candidate identifier corresponds to the first label,
wherein the first machine learning model is trained to predict the predicted label probability corresponding to the first label based at least in part on (1) a label probability set and (2) a first data subset of a candidate data set associated with a candidate pool identifier set,
wherein the first data subset is based at least in part on:
(1) a first candidate identifier subset corresponding to the first label, and
(2) a long-term record threshold,
wherein the label probability set is identified using a second machine learning model based at least in part on a second data subset of the candidate data set,
and wherein the second data subset is based at least in part on:
(1) the first candidate identifier subset corresponding to the first label, and
(2) a short-term record threshold.

6 more elements

A computing system comprising one or more processors and non-transitory memory having computer-coded instructions stored thereon that, in execution with the one or more processors, causes the computing system to:

generate, using a first machine learning model, a predicted label probability corresponding to a first label, wherein the predicted label probability indicates a probability that a candidate identifier corresponds to the first label,
wherein the first machine learning model is trained to predict the predicted label probability corresponding to the first label based at least in part on (1) a label probability set and (2) a first data subset of a candidate data set associated with a candidate pool identifier set,
wherein the first data subset is based at least in part on:
(1) a first candidate identifier subset corresponding to the first label, and
(2) a long-term record threshold,
wherein the label probability set is identified using a second machine learning model based at least in part on a second data subset of the candidate data set,
and wherein the second data subset is based at least in part on:
(1) the first candidate identifier subset corresponding to the first label, and
(2) a short-term record threshold.

6 more elements

A computer program product comprising a non-transitory computer-readable storage medium having computer program code stored thereon that, in execution with one or more processors, configures the computer program product for:

generating, using a first machine learning model, a predicted label probability corresponding to a first label, wherein the predicted label probability indicates a probability that a candidate identifier corresponds to the first label,
wherein the first machine learning model is trained to predict the predicted label probability corresponding to the first label based at least in part on (1) a label probability set and (2) a first data subset of a candidate data set associated with a candidate pool identifier set,
wherein the first data subset is based at least in part on:
(1) a first candidate identifier subset corresponding to the first label, and
(2) a long-term record threshold,
wherein the label probability set is identified using a second machine learning model based at least in part on a second data subset of the candidate data set,
and wherein the second data subset is based at least in part on:
(1) the first candidate identifier subset corresponding to the first label, and
(2) a short-term record threshold.

6 more elements

Dependencies are inferred from the claim text.

Apparatus, computer program product, and method for predictive data labelling using a dual-prediction model system

Life of this patent

Apparatus, computer program product, and method for predictive data labelling using a dual-prediction model system

Life of this patent

Abstract

Claims20