Back to results
US 11,698,934 B2Granted
Graph-embedding-based paragraph vector machine learning models
AssigneeOptum, Inc.
InventorsSuman Roy, Amit Kumar, Ayan Sengupta, Riccardo Mattivi, Ahmed Selim, Shashi Kumar
Life of this patent
Estimating the term…
Abstract
Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive structural analysis on document data objects that are associated with an ontology graph. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive data analysis operations on document data objects that are associated with an ontology graph using document embeddings that are generated by graph-embedding-based paragraph vector machine learning models.
Claims21
A computer-implemented method comprising:
- identifying, by one or more processors, a plurality of word vectors associated with a document data object characterized by an ontology graph;
- identifying, by the one or more processors, one or more relationships of the document data object based at least in part on the ontology graph;
- generating, by the one or more processors and using a graph-embedding-based paragraph vector machine learning model, a document representation and one or more relational representations for the one or more relationships based at least in part on the plurality of word vectors and the one or more relationships, wherein:
- the graph-embedding-based paragraph vector machine learning model is configured to optimize a textual-relational optimization output that comprises a textual optimization sub-output and a relational optimization sub-output,
- the textual optimization sub-output is generated based at least in part on (a) the plurality of word vectors, (b) the document representation, and (c) a weighted distance value for a secondary document data object of a plurality of document data objects other than the document data object,
- the weighted distance value is determined based at least in part on (a) a distance value between the document data object and the secondary document data object and (b) a relational representation for a relationship type of the secondary document data object and the document data object, and
- the relational optimization sub-output is generated based at least in part on the document representation and the one or more relational representations;
- and
- initiating, by the one or more processors, the performance of one or more prediction-based actions based at least in part on at least one of the document representation or the one or more relational representations.
6 more elements
An apparatus comprising one or more processors and memory including program code, the memory and the program code configured to, with the one or more processors, cause the apparatus to at least:
- identify a plurality of word vectors associated with a document data object characterized by an ontology graph;
- identify one or more relationships of the document data object based at least in part on the ontology graph;
- generate, using a graph-embedding-based paragraph vector machine learning model, a document representation and one or more relational representations for the one or more relationships based at least in part on the plurality of word vectors and the one or more relationships, wherein:
- the graph-embedding-based paragraph vector machine learning model is configured to optimize a textual-relational optimization output that comprises a textual optimization sub-output and a relational optimization sub-output,
- the textual optimization sub-output is generated based at least in part on (a) the plurality of word vectors, (b) the document representation, and (c) a weighted distance value for a secondary document data object of a plurality of document data objects other than the document data object,
- the weighted distance value is determined based at least in part on (a) a distance value between the document data object and the secondary document data object and (b) a relational representation for a relationship type of the secondary document data object and the document data object, and
- the relational optimization sub-output is generated based at least in part on the document representation and the one or more relational representations;
- and
- perform one or more prediction-based actions based at least in part on at least one of the document representation and the one or more relational representations.
6 more elements
A computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions, when executed by one or more processors, cause the one or more processors to:
- identify a plurality of word vectors associated with a document data object characterized by an ontology graph;
- identify one or more relationships of the document data object based at least in part on the ontology graph;
- generate, using a graph-embedding-based paragraph vector machine learning model, a document representation and one or more relational representations for the one or more relationships based at least in part on the plurality of word vectors and the one or more relationships, wherein:
- the graph-embedding-based paragraph vector machine learning model is configured to optimize a textual-relational optimization output that comprises a textual optimization sub-output and a relational optimization sub-output,
- the textual optimization sub-output is generated based at least in part on (a) the plurality of word vectors, (b) the document representation, and (c) a weighted distance value for a secondary document data object of a plurality of document data objects other than the document data object,
- the weighted distance value is determined based at least in part on (a) a distance value between the document data object and the secondary document data object and (b) a relational representation for a relationship type of the secondary document data object and the document data object, and
- the relational optimization sub-output is generated based at least in part on the document representation and the one or more relational representations;
- and
- perform one or more prediction-based actions based at least in part on at least one of the document representation and the one or more relational representations.
6 more elements
Dependencies are inferred from the claim text.