US 12,596,936 B2Granted

Predictive data analysis techniques using graph-based code recommendation machine learning models

AssigneeOptum Technology, Inc.

InventorsNilav Baran Ghosh, Abhilash Sivva, Srikanth B. Adibhatla

Life of this patent

Estimating the term…

Loading patent…

Abstract

Solutions for more efficient and effective predictive code recommendation are disclosed. In one example, a method includes identifying a graph-based code recommendation machine learning model, wherein each inferred edge weight value of the graph-based code recommendation machine learning model is updated based at least in part on each compressed forward-adjusted temporal distance measure for an observed co-occurrence of any observed co-occurrences of a predictive code pair for the inferred edge weight value within one or more temporally-proximate occurrence subsets determined based at least in part on a plurality of training predictive code occurrences; processing the input predictive code using the graph-based code recommendation machine learning model to generate one or more related codes of the plurality of predictive codes for the input predictive code; and performing one or more prediction-based actions based at least in part on the one or more related codes.

Claims20

Dependent claims

A computer-implemented method comprising:

receiving, by one or more processors, a request identifying an input predictive code of a plurality of predictive codes;
determining, by the one or more processors, using a graph-based machine learning model and based at least in part on the input predictive code, a related code for the input predictive code by:
(i) determining, using the graph-based machine learning model, a first predictive code node corresponding to the input predictive code, wherein the graph-based machine learning model comprises the first predictive code node associated with the input predictive code and a second predictive code node associated with the related code, and
(ii) determining, using the graph-based machine learning model, the second predictive code node is related to the first predictive code node based at least in part on an inferred edge weight value associated with an edge between the second predictive code node and the first predictive code node, wherein:
(a) the first predictive code node is associated with a set of predictive code occurrences for the input predictive code and a set of occurrence timestamps respectively associated with the set of predictive code occurrences;
(b) the inferred edge weight value is based at least in part on a compressed forward-adjusted temporal distance measure between a first occurrence timestamp corresponding to a first predictive code occurrence of the input predictive code and a second occurrence timestamp corresponding to a second predictive code occurrence of the related code;
and
(c) the graph-based machine learning model is trained using a set of training predictive codes within a temporally-proximate occurrence subset, wherein (i) the temporally-proximate occurrence subset comprises a group of predictive code occurrences within a temporal range, (ii) the temporally-proximate occurrence subset comprises a predictive code pair, and (iii) the graph-based machine learning model is trained based at least in part on a pairwise temporal distance measure associated with the predictive code pair;
and
providing for display, by the one or more processors, the related code.

7 more elements

A system comprising one or more processors and at least one memory storing processor executable instructions that, when executed by the one or more processors, cause the one or more processors to:

receive a request identifying an input predictive code of a plurality of predictive codes;
determine, using a graph-based machine learning model, and based at least in part on the input predictive code, a related code for the input predictive code by:
(i) determining, using the graph-based machine learning model comprises a first predictive code node corresponding to the input predictive code, wherein the graph-based machine learning model comprises the first predictive code node associated with the input predictive code and a second predictive code node associated with the related code, and
(ii) determining, using the graph-based machine learning model, the second predictive code node is related to the first predictive code node based at least in part on an inferred edge weight value associated with an edge between the second predictive code node and the first predictive code node, wherein:
(a) the first predictive code node a set of predictive code occurrences for the input predictive code and a set of occurrence timestamps respectively associated with the set of predictive code occurrences;
(b) the inferred edge weight value is based at least in part on a compressed forward-adjusted temporal distance measure between a first occurrence timestamp corresponding to a first predictive code occurrence of the input predictive code and a second occurrence timestamp corresponding to a second predictive code occurrence of the related code;
and
(c) the graph-based machine learning model is trained using a set of training predictive codes within a temporally-proximate occurrence subset, wherein (i) the temporally-proximate occurrence subset comprises a group of predictive code occurrences within a temporal range, (ii) the temporally-proximate occurrence subset comprises a predictive code pair, and (iii) the graph-based machine learning model is trained based at least in part on a pairwise temporal distance measure associated with the predictive code pair;
and
provide for display the related code.

7 more elements

One or more non-transitory computer-readable storage media including instructions that, when executed by one or more processors, cause the one or more processors to:

receive a request identifying an input predictive code of a plurality of predictive codes;
determine, using a graph-based machine learning model, and based at least in part on the input predictive code, a related code for the input predictive code by
(i) determining, using the graph-based machine learning model, a first predictive code node corresponding to the input predictive code, wherein the graph-based machine learning model comprises the first predictive code node associated with the input predictive code and a second predictive code node associated with the related code, and
(ii) determining, using the graph-based machine learning model, the second predictive code node is related to the first predictive code node based at least in part on an inferred edge weight value associated with an edge between the second predictive code node and the first predictive code node, wherein:
(a) the first predictive code node is associated with a set of predictive code occurrences for the input predictive code and a set of occurrence timestamps respectively associated with the set of predictive code occurrences;
(b) the inferred edge weight value is based at least in part on a compressed forward-adjusted temporal distance measure between a first occurrence timestamp associated with a first predictive code occurrence of the input predictive code and a second occurrence timestamp associated with a second predictive code occurrence of the related code;
and
(c) the graph-based machine learning model is trained using a set of training predictive codes within a temporally-proximate occurrence subset, wherein (i) the temporally-proximate occurrence subset comprises a group of predictive code occurrences within a defined temporal range, (ii) temporally-proximate occurrence subset comprises a predictive code pair, and (iii) the graph-based machine learning model is trained based at least in part on a pairwise temporal distance measure associated with the predictive code pair;
and
provide for display the related code.

7 more elements

Dependencies are inferred from the claim text.