Method and system for recommending promotions to consumers

One or more non-transitory computer-readable media having encoded thereon one or more computer-executable instructions for performing a method for associating a plurality of consumers using a promotion and marketing service, the method comprising:

• programmatically retrieving one or more profile data features associated with a first consumer;
• programmatically retrieving one or more profile data features associated with a second consumer;
• programmatically retrieving data on one or more prior activities performed by the first consumer and the second consumer in relation to one or more prior promotions offered to the first consumer and the second consumer, wherein the data on one or more prior activities is generated based at least on clickstream data received from a first input at a first graphical user interface of a first consumer device associated with the first consumer and a second input at a second graphical user interface of a second consumer device associated with the second consumer;
• automatically generating an association between the first consumer and the second consumer, wherein generating the association between the first consumer and the second consumer comprises:
• analyzing a social graph to identify a connection path connecting a first node representing the first consumer and a second node representing the second consumer;
• retrieving a connection score for the connection path, the connection score representing an estimated strength of connection between the first consumer and the second consumer, the connection score based on the one or more prior activities performed by the first consumer in relation to the one or more prior promotions and the one or more prior activities performed by the second consumer in relation to the one or more prior promotions;
• analyzing a taste graph to identify a path connecting a first node representing the first consumer and a second node representing the second consumer;
• retrieving a relevancy score for the path, the relevancy score representing an estimated relevancy of promotion interests between the first consumer and the second consumer, the relevancy score based on the one or more profile data features associated with the first consumer and the one or more profile data features associated with the second consumer;
• generating, via a machine learning model, an association score related to the association between the first consumer and the second consumer based on the connection score and the relevancy score, wherein generating the association score via the machine learning model comprises generating a multidimensional vector for the first consumer and the second consumer and comparing the multidimensional vector for the first consumer and the multidimensional vector for the second consumer to determine a measure of similarity and programmatically executing a machine learning algorithm to relate the measure of similarity to a probability that the association score between the first consumer and the second consumer is positive or negative;
• and
• generating the association between the first consumer and the second consumer in response to the association score exceeding a predetermined threshold score;
• and
• recording the association between the first consumer and the second consumer in a consumer association database stored on the one or more non-transitory computer-readable medium.

A system comprising:

• a storage device configured to store a consumer association database;
• and
• processing circuitry configured to:
• programmatically retrieve one or more profile data features associated with a first consumer;
• programmatically retrieving one or more profile data features associated with a second consumer;
• programmatically retrieve data on one or more prior activities performed by the first consumer and the second consumer in relation to one or more prior promotions offered to the first consumer and the second consumer, wherein the data on one or more prior activities is generated based at least on clickstream data received from a first input at a first graphical user interface of a first consumer device associated with the first consumer and a second input at a second graphical user interface of a second consumer device associated with the second consumer;
• automatically generate, using circuitry provided on a computing device, an association between the first consumer and the second consumer, wherein generating the association between the first consumer and the second consumer comprises:
• analyzing a social graph to identify a connection path connecting a first node representing the first consumer and a second node representing the second consumer;
• retrieving a connection score for the connection path, the connection score representing an estimated strength of connection between the first consumer and the second consumer, the connection score based on the one or more prior activities performed by the first consumer in relation to the one or more prior promotions and the one or more prior activities performed by the second consumer in relation to the one or more prior promotions;
• analyzing a taste graph to identify a path connecting a first node representing the first consumer and a second node representing the second consumer;
• retrieving a relevancy score for the path, the relevancy score representing an estimated relevancy of promotion interests between the first consumer and the second consumer, the relevancy score based on the one or more profile data features associated with the first consumer and the one or more profile data features associated with the second consumer;
• generating, via a machine learning model, an association score related to the association between the first consumer and the second consumer based on the connection score and the relevancy score, wherein generating the association score via the machine learning model comprises generating a multidimensional vector for the first consumer and the second consumer and comparing the multidimensional vector for the first consumer and the multidimensional vector for the second consumer to determine a measure of similarity and programmatically executing a machine learning algorithm to relate the measure of similarity to a probability that the association score between the first consumer and the second consumer is positive or negative;
• and
• generating the association between the first consumer and the second consumer in response to the association score exceeding a predetermined threshold score;
• and
• record the association between the first consumer and the second consumer in a consumer association database stored on a non-transitory computer-readable medium.