Back to results
US 12,511,664 B2Granted
System and method for automatically optimizing customer communication
AssigneeGoogle Inc.
InventorsDimitris Meretakis, Zigmars Rasscevskis, Vinsensius B. Vega S. Naryanto, Tom Beyer, Szabolcs Payrits, Martin Stolle, Mark Steven Schadler, Jack Willow Waldron, Ali Galip Bayrak
Life of this patent
Estimating the term…
Abstract
The present disclosure provides a closed loop, self-learning system that automatically optimizes what experiences should be presented to each customer. Instead of relying on rules and external targeting, it observes customer reactions to continuously improve performance and adapt to environment changes.
Claims19
A computer-implemented method of communication, comprising:
- receiving data relating to an institution and its customer, the data including a specified goal for the institution;
- training a machine learning model using the received data, the machine learning model comprising a dynamic next best action engine;
- determining a context for the customer at any given time;
- dynamically selecting, using the dynamic next best action engine, one or more experiences for delivery to the customer, wherein the selecting of the one or more experiences is optimized based on the determined context and the specified goal;
- delivering the one or more experiences to the customer;
- identifying, using a measurement service, one or more customer actions, the identifying comprising capturing interactions in an event log;
- providing the event log to the dynamic next best action engine;
- determining a value relating to the one or more experiences based on the one or more customer actions, the value quantifying a benefit to the institution of delivering the one or more experiences, the value being associated with the customer;
- and
- continuously updating the machine learning model based on the interactions and using the determined value to improve selection of future experiences.
7 more elements
A system, comprising:
- one or more memories;
- and
- one or more processors in communication with the one or more memories, the one or more processors configured to:
- receive data relating to an institution and its customer, the data including a specified goal for the institution;
- train a machine learning model using the received data, the machine learning model comprising a dynamic next best action engine;
- determine a context for the customer at a given time;
- dynamically select, using the dynamic next best action engine, one or more experiences for delivery to the customer, wherein the selecting of the one or more experiences is optimized based on the determined context and the specified goal;
- deliver the one or more experiences to the customer;
- identify one or more customer actions, the identifying comprising capturing interactions in an event log;
- provide the event log to the dynamic next best action engine;
- determine a value relating to the one or more experiences based on the one or more customer actions, the value quantifying a benefit to the institution of delivering the one or more experiences, the value being associated with the customer;
- and
- continuously update the machine learning model based on the interactions and using the determined value to improve selection of future experiences.
10 more elements
A non-transitory computer-readable medium storing instructions executable by one or more processors for performing a method of communication, comprising:
- receiving data relating to an institution and its customer, the data including a specified goal for the institution;
- training a machine learning model using the received data, the machine learning model comprising a dynamic next best action engine;
- determining a context for the customer at a given time;
- dynamically selecting, using the dynamic next best action engine, one or more experiences for delivery to the customer, wherein the selecting of the one or more experiences is optimized based on the determined context and the specified goal;
- delivering the one or more experiences to the customer;
- identifying one or more customer actions, the identifying comprising capturing interactions in an event log;
- providing the event log to the dynamic next best action engine;
- determining a value relating to the one or more experiences based on the one or more customer actions, the value quantifying a benefit to the institution of delivering the one or more experiences, the value being associated with the customer;
- and
- continuously updating the machine learning model based on the interactions and using the determined value to improve selection of future experiences.
7 more elements
Dependencies are inferred from the claim text.