US 11,455,347 B2Granted

Dynamic detection of cross-document associations

AssigneeOptum Technology, Inc.

InventorsSwapna Sourav Rout, Sharlene L. Tan, Vamsi Bhandaru, Vineet Shukla, Akshaya D. N, Kartik Chaudhary

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods are configured to generate a set of related document objects for a predictive entity and/or to generate an optimal document sequence for a set of related document objects. In one embodiment, a set of related document objects for a predictive entity is generated by processing entity metadata features associated with the predictive entity using an entity-document correlation machine learning model, and an optimal document sequence is generated for the set of related document objects by processing the set of related document objects using a document sequence optimization machine learning model.

Claims20

Dependent claims

A computer-implemented method for generating an optimal document sequence for a set of input document objects, the computer-implemented method comprising:

for each input document object in the set of input document objects, determining, using one or more processors, a set of sequence-related features;
processing sets of sequence-related features for the set of input document objects using the one or more processors and a document sequence optimization machine learning model to generate the optimal document sequence, wherein:
(a) the document sequence optimization machine learning model is configured to (i) generate a set of candidate document sequences for the set of input document objects, (ii) generate a combined sequence probability value for each candidate document sequence of the set of candidate document sequences, and (iii) select the optimal document sequence from the set of candidate document sequences, and (b) generating the combined sequence probability value for a particular candidate document sequence comprises:
generating, using the one or more processors, an individual occurrence probability for an initial input document object of the set of input documents that is associated with a lower-most document ordering value of one or more document sequence ordering values defined by the candidate document sequence;
for each current document ordering value of the one or more document sequence ordering values other than the lower-most document ordering value, determining, using the one or more processors, a conditional occurrence probability for occurrence of a post-initial input document object of the set of input document objects that is associated with the current document ordering value following each input document object of the set of input document objects having a lower document ordering value relative to the current document ordering value;
and
generating, using the one or more processors, the combined sequence probability value based at least in part on the individual occurrence probability for the initial input document object and each conditional occurrence probability for a post-initial input document object of the set of input document objects;
and
performing, using the one or more processors, one or more prediction-based actions based at least in part on the optimal document sequence.

6 more elements

An apparatus for generating an optimal document sequence for a set of input document objects, the apparatus comprising at least one processor and at least one non-transitory memory including program code, the at least one non-transitory memory and the program code configured to, with the processor, cause the apparatus to at least:

for each input document object in the set of input document objects, determine a set of sequence-related features;
process sets of sequence-related features for the set of input document objects using a document sequence optimization machine learning model to generate the optimal document sequence, wherein:
(a) the document sequence optimization machine learning model is configured to (i) generate a set of candidate document sequences for the set of input document objects, (ii) generate a combined sequence probability value for each candidate document sequence of the set of candidate document sequences, and (iii) select the optimal document sequence from the set of candidate document sequences, and (b) generating the combined sequence probability value for a particular candidate document sequence comprises:
generating an individual occurrence probability for an initial input document object of the set of input documents that is associated with a lower-most document ordering value of one or more document sequence ordering values defined by the candidate document sequence;
for each current document ordering value of the one or more document sequence ordering values other than the lower-most document ordering value, determining a conditional occurrence probability for occurrence of a post-initial input document object of the set of input document objects that is associated with the current document ordering value following each input document object of the set of input document objects having a lower document ordering value relative to the current document ordering value;
and
generating the combined sequence probability value based at least in part on the individual occurrence probability for the initial input document object and each conditional occurrence probability for a post-initial input document object of the set of input document objects;
and
perform one or more prediction-based actions based at least in part on the optimal document sequence.

6 more elements

A non-transitory computer storage medium comprising instructions for generating an optimal document sequence for a set of input document objects, the instructions being configured to cause one or more processors to at least perform operations configured to:

for each input document object in the set of input document objects, determine a set of sequence-related features;
process sets of sequence-related features for the set of input document objects using a document sequence optimization machine learning model to generate the optimal document sequence, wherein:
(a) the document sequence optimization machine learning model is configured to (i) generate a set of candidate document sequences for the set of input document objects, (ii) generate a combined sequence probability value for each candidate document sequence of the set of candidate document sequences, and (iii) select the optimal document sequence from the set of candidate document sequences, and (b) generating the combined sequence probability value for a particular candidate document sequence comprises:
generating an individual occurrence probability for an initial input document object of the set of input documents that is associated with a lower-most document ordering value of one or more document sequence ordering values defined by the candidate document sequence;
for each current document ordering value of the one or more document sequence ordering values other than the lower-most document ordering value, determining a conditional occurrence probability for occurrence of a post-initial input document object of the set of input document objects that is associated with the current document ordering value following each input document object of the set of input document objects having a lower document ordering value relative to the current document ordering value;
and
generating the combined sequence probability value based at least in part on the individual occurrence probability for the initial input document object and each conditional occurrence probability for a post-initial input document object of the set of input document objects;
and
perform one or more prediction-based actions based at least in part on the optimal document sequence.

6 more elements

Dependencies are inferred from the claim text.