Machine learning based user targeting

A method, comprising:

• receiving, by one or more processing devices, subject data associated with a subject, the subject data comprising data that describes the subject;
• obtaining, by the one or more processing devices, external third party entity data from at least one online source and supplementing the external third party entity data with domain-specific entity data comprising internal entity data to create a customized entity data set;
• training, by the one or more processing devices, a machine learning model using the customized entity data set;
• determining, by the one or more processing devices, using the machine learning model, an affinity level of each of one or more entities in relation to the subject, the subject data provided as input to the machine learning model, the machine learning model providing output for determining the affinity level based at least in part on a digital encoding of semantic similarity between an entity and the subject using text descriptions of the entity and the subject, the affinity level indicating the entity's likelihood of providing a response related to the subject;
• generating an interactive graphical interface comprising graphical representations of one or more entities, the one or more entities divided into segmented groups within the graphical interface according to their affinity levels with the subject such that each entity within a segmentation group is indicated with a same graphical representation that is different from a graphical representation of a different segmentation group;
• presenting the interactive graphical interface on a display;
• receiving an input on one of the graphical representations of an entity within the graphical interface in response to detecting a user's interaction with the one of the graphical representations;
• dynamically presenting metadata for an entity in response to receiving input on the graphical representation of the entity;
• generating, by the one or more processing devices, one or more promotional campaign strategies for the subject based at least in part on the affinity level of each of the one or more entities in relation to the subject, the one or more promotional campaign strategies targeting the subject to the one or more entities;
• monitoring for new entity data from the at least one online source and supplements the customized entity data set with new entity data that is available from the at least one online source;
• receiving, by the one or more processing devices, feedback associated with the one or more promotional campaign strategies;
• and
• iteratively retraining, by the one or more processing devices, the machine learning model using the received feedback and the supplemented customized entity data to refine the one or more promotional campaign strategies for the subject.

A computer program product, comprising a computer-readable storage medium having program code embodied therein, the program code executable by one or more processing devices for:

• receiving subject data associated with a subject, the subject data comprising data that describes the subject;
• obtaining external third party entity data from at least one online source and supplementing the external third party entity data with domain-specific entity data comprising internal entity data to create a customized entity data set;
• training a machine learning model using the customized entity data set;
• determining, using the machine learning model, an affinity level of each of one or more entities in relation to the subject, the subject data provided as input to the machine learning model, the machine learning model providing output for determining the affinity level based at least in part on a digital encoding of semantic similarity between an entity and the subject using text descriptions of the entity and the subject, the affinity level indicating the entity's likelihood of providing a response related to the subject;
• generating an interactive graphical interface comprising graphical representations of one or more entities, the one or more entities divided into segmented groups within the graphical interface according to their affinity levels with the subject such that each entity within a segmentation group is indicated with a same graphical representation that is different from a graphical representation of a different segmentation group;
• presenting the interactive graphical interface on a display;
• receiving an input on one of the graphical representations of an entity within the graphical interface in response to detecting a user's interaction with the one of the graphical representations;
• dynamically presenting metadata for an entity in response to receiving input on the graphical representation of the entity;
• generating one or more promotional campaign strategies for the subject based at least in part on the affinity level of each of the one or more entities in relation to the subject, the one or more promotional campaign strategies targeting the subject to the one or more entities;
• monitoring for new entity data from the at least one online source and supplements the customized entity data set with new entity data that is available from the at least one online source;
• receiving feedback associated with the one or more promotional campaign strategies;
• iteratively retraining the machine learning model using the received feedback and the supplemented customized entity data to refine the one or more promotional campaign strategies for the subject;
• and
• outputting, for display, a visual representation of the affinity level of each of the one or more entities in relation to the subject.

A method, comprising:

• receiving, by one or more processing devices, subject data associated with a subject, the subject data comprising data that describes the subject;
• obtaining, by the one or more processing devices, external third party entity data from at least one online source and supplementing the external third party entity data with domain-specific entity data comprising internal entity data to create a customized entity data set;
• training, by the one or more processing devices, a machine learning model using the customized entity data set;
• determining, by the one or more processing devices, using the machine learning model, an affinity level of each of one or more entities in relation to the subject, the subject data provided as input to the machine learning model, the machine learning model providing output for determining the affinity level based at least in part on a digital encoding of semantic similarity between an entity and the subject using text descriptions of the entity and the subject, the affinity level indicating the entity's likelihood of providing a response related to the subject;
• generating an interactive graphical interface comprising graphical representations of one or more entities, the one or more entities divided into segmented groups within the graphical interface according to their affinity levels with the subject such that each entity within a segmentation group is indicated with a same graphical representation that is different from a graphical representation of a different segmentation group;
• presenting the interactive graphical interface on a display;
• receiving an input on one of the graphical representations of an entity within the graphical interface in response to detecting a user's interaction with the one of the graphical representations;
• dynamically presenting metadata for an entity in response to receiving input on the graphical representation of the entity;
• determining one or more messaging strategies for each of the one or more entities based on the affinity levels of each of the one or more entities;
• presenting a list of messaging strategies for each of the one or more entities in an order ranked by a determined messaging effectiveness;
• and
• dynamically generating a modifiable messaging template based on a selected one of the list of messaging strategies for a selected entity;
• generating, by the one or more processing devices, one or more promotional campaign strategies for the subject based at least in part on the affinity level of each of the one or more entities in relation to the subject, the one or more promotional campaign strategies targeting the subject to the one or more entities;
• monitoring for new entity data from the at least one online source and supplements the customized entity data set with new entity data that is available from the at least one online source;
• receiving, by the one or more processing devices, feedback associated with the one or more promotional campaign strategies;
• iteratively retraining, by the one or more processing devices, the machine learning model using the received feedback and the supplemented customized entity data to refine the one or more promotional campaign strategies for the subject;
• and
• outputting, for display, by the one or more processing devices, a visual representation of the affinity level of each of the one or more entities in relation to the subject.