US 12,683,893 B2Granted

Routing messages based on message priority

AssigneeCapital One Services, LLC

InventorsJacob Berger, Charles DeVore, Xiaoyi Wang, Lindsey May, Waleed Algadhi, Linsu Han

Life of this patent

Estimating the term…

Loading patent…

Abstract

Methods and systems for routing messages based on message priority. In some aspects, the system may determine paths between a source and one or more destinations. The system may determine, for a first message, using a machine learning model, a first predicted path having a first distance. The system may determine, based on the first predicted path, using a Markov decision process, a priority of the first message. Based on determining that a first confidence score for the first predicted path does not exceed a threshold, the system may select a different path for the first message. The system may select the different path based on the priority of the first message and a second distance associated with the different path. The system may then re-direct the first message along the different path.

Claims20

Dependent claims

A system for routing messages based on priorities, the system comprising:

one or more processors and one or more non-transitory computer-readable media having computer-executable instructions stored thereon, the computer-executable instructions, when executed by the one or more processors, causing operations comprising:
determining a plurality of paths for routing messages from a source to one or more destinations, wherein each path of the plurality of paths is associated with a respective distance between the source and a respective destination of the one or more destinations;
determining, for a first message of a plurality of messages, using a machine learning model, a first predicted path of the plurality of paths, the first predicted path connecting the source and a first predicted destination of the one or more destinations;
determining, based on the first predicted path, a priority of the first message using a Markov decision process, wherein the Markov decision process utilizes agent input and a probabilistic determination;
determining whether a first confidence score associated with selecting the first predicted path for the first message exceeds a threshold, wherein the first confidence score is determined based on the priority of the first message, a first distance associated with the first predicted path, and the first predicted destination;
and
based on determining that the first confidence score for the first predicted path does not exceed the threshold, selecting a different path of the plurality of paths for the first message, the different path connecting the source and a second predicted destination of the one or more destinations, wherein the different path is selected based on the priority of the first message, a second distance associated with the different path, and the second predicted destination, wherein a second confidence score for the different path exceeds the threshold.

4 more elements

A method comprising:

determining a plurality of paths between a source and one or more destinations, wherein each path of the plurality of paths is associated with a respective distance between the source and a respective destination of the one or more destinations;
determining, for a first candidate of a plurality of candidates, using a first model, a first predicted path of the plurality of paths;
determining, based on the first predicted path, using a second model, a priority of the first candidate using agent input and a probabilistic determination;
determining that a first confidence score associated with selecting the first predicted path for the first candidate does not exceed a threshold, wherein the first confidence score is determined based on the priority of the first candidate and a first distance associated with the first predicted path;
and
based on determining that the first confidence score for the first predicted path does not exceed the threshold, selecting a different path of the plurality of paths for the first candidate, wherein the different path is selected based on the priority of the first candidate and a second distance associated with the different path, wherein a second confidence score for the different path exceeds the threshold.

3 more elements

One or more non-transitory, computer-readable media storing instructions that, when executed by one or more processors, cause operations comprising:

determining a plurality of paths between a source and one or more destinations, wherein each path of the plurality of paths is associated with a respective distance between the source and a respective destination of the one or more destinations;
determining, for a first candidate of a plurality of candidates, using a first model, a first predicted path of the plurality of paths;
determining, based on the first predicted path, using a second model, a priority of the first candidate using agent input and a probabilistic determination;
determining that a first confidence score associated with selecting the first predicted path for the first candidate does not exceed a threshold;
and
based on determining that the first confidence score for the first predicted path does not exceed the threshold, selecting a different path of the plurality of paths for the first candidate, wherein a second confidence score for the different path exceeds the threshold.

3 more elements

Dependencies are inferred from the claim text.