Back to results
US 12,634,395 B2Granted
Routing of communications to contact center agents using machine learning
AssigneeCisco Technology, Inc.
InventorsArunabh Bhattacharjee, Aseem Banshidhar Asthana
Life of this patent
Estimating the term…
Abstract
In one example embodiment, one or more machine learning models of at least one processor determine an agent of a communication center to receive a communication from a user according to one or more objectives that optimize a corresponding metric from interaction of the user and the agent. Features for training the one or more machine learning models are dynamically selected based on relevance of the features to attaining the one or more objectives. The at least one processor routes the communication to the agent of the communication center.
Claims20
A method comprising:
- determining, via at least one processor, an applicable rule for routing a first communication from among preconfigured rules based on occurrence of conditions specified in the preconfigured rules;
- routing, via the at least one processor, the first communication to a first agent of a communication center based on one or more agents indicated by the applicable rule;
- determining, via the at least one processor, an absence of a rule in the preconfigured rules for a second communication;
- determining, via a plurality of machine learning models of the at least one processor, second agents of the communication center to receive the second communication according to a plurality of objectives that each optimize a corresponding metric from agent interaction, wherein the plurality of machine learning models are each trained for an associated objective and further determine probabilities representing confidences for the second agents;
- producing, via the at least one processor, scores for the second agents by applying a weight for the associated objective of a corresponding machine learning model to a probability for a corresponding second agent determined by the corresponding machine learning model;
- determining, via the at least one processor, a second agent from among the second agents to receive the second communication based on the scores for the second agents;
- routing, via the at least one processor, the second communication to the second agent of the communication center;
- training, via the at least one processor, a plurality of different types of machine learning models for a corresponding objective, wherein the plurality of different types of machine learning models include a random forest classifier, a linear support vector classifier, a Multinomial Naive Bayes model, a logistic regression model, and a gradient boosting classifier;
- and
- replacing, via the at least one processor, a machine learning model of a first type of the plurality of machine learning models associated with the corresponding objective with a new machine learning model of a second different type selected from the plurality of different types of machine learning models and having greater accuracy than the machine learning model of the first type.
7 more elements
An apparatus comprising:
- a communication center comprising one or more processors, wherein the one or more processors are configured to perform operations including:
- determining a rule for routing a communication from among preconfigured rules based on occurrence of conditions specified in the preconfigured rules;
- in response to a presence of the rule in the preconfigured rules, determining an agent of the communication center to receive the communication based on one or more agents indicated by the rule;
- in response to an absence of the rule in the preconfigured rules:
- determining, via a plurality of machine learning models, a plurality of agents of the communication center to receive the communication according to a plurality of objectives that each optimize a corresponding metric from agent interaction, wherein the plurality of machine learning models are each trained for an associated objective and further determine probabilities representing confidences for the plurality of agents;
- producing scores for the plurality of agents by applying a weight for the associated objective of a corresponding machine learning model to a probability for a corresponding agent determined by the corresponding machine learning model;
- and
- determining the agent from among the plurality of agents to receive the communication based on the scores for the plurality of agents;
- routing the communication to the agent of the communication center;
- training a plurality of different types of machine learning models for a corresponding objective, wherein the plurality of different types of machine learning models include a random forest classifier, a linear support vector classifier, a Multinomial Naive Bayes model, a logistic regression model, and a gradient boosting classifier;
- and
- replacing a machine learning model of a first type of the plurality of machine learning models associated with the corresponding objective with a new machine learning model of a second different type selected from the plurality of different types of machine learning models and having greater accuracy than the machine learning model of the first type.
9 more elements
One or more non-transitory computer readable storage media encoded with processing instructions that, when executed by one or more processors, cause the one or more processors to perform operations including:
- determining a rule for routing a communication from among preconfigured rules based on occurrence of conditions specified in the preconfigured rules;
- in response to a presence of the rule in the preconfigured rules, determining an agent of a communication center to receive the communication based on one or more agents indicated by the rule;
- in response to an absence of the rule in the preconfigured rules:
- determining, via a plurality of machine learning models, a plurality of agents of the communication center to receive the communication according to a plurality of objectives that each optimize a corresponding metric from agent interaction, wherein the plurality of machine learning models are each trained for an associated objective and further determine probabilities representing confidences for the plurality of agents;
- producing scores for the plurality of agents by applying a weight for the associated objective of a corresponding machine learning model to a probability for a corresponding agent determined by the corresponding machine learning model;
- and
- determining the agent from among the plurality of agents to receive the communication based on the scores for the plurality of agents;
- routing the communication to the agent of the communication center;
- training a plurality of different types of machine learning models for a corresponding objective, wherein the plurality of different types of machine learning models include a random forest classifier, a linear support vector classifier, a Multinomial Naive Bayes model, a logistic regression model, and a gradient boosting classifier;
- and
- replacing a machine learning model of a first type of the plurality of machine learning models associated with the corresponding objective with a new machine learning model of a second different type selected from the plurality of different types of machine learning models and having greater accuracy than the machine learning model of the first type.
8 more elements
Dependencies are inferred from the claim text.