System, method, and computer program for providing an instance of a promotional message to a user based on a predicted emotional response corresponding to user characteristics

A non-transitory computer-readable medium comprising a computer program, that, when executed by a computer system, enables the computer system to perform the following method for providing a promotional message to a user based on a predicted emotional response, the method comprising:

• creating a statistical model that, for each of a set of emotions, predicts the probability that a user with certain characteristics will respond to the emotion, wherein creating the model comprises:
• receiving training data related to a plurality of training users, wherein, for each training user, the training data received includes user characteristics, a plurality of promotional messages provided to the training user, and, for each of the messages, an indication of whether the training user responded to the message;
• automatically tagging each of the promotional messages in the training data with one or more emotions to identify the emotions received by the training users;
• and
• building the statistical model using a statistical machine learning algorithm, wherein the model is trained using the user characteristics of the training users, the emotions provided to the training users, and the corresponding response data;
• using machine-learning-based modeling to automatically predict the emotion to which a non-training user is most likely to respond and to use the prediction to provide a promotional message to the non-training user by performing the following:
• receiving user characteristics for a non-training user;
• creating a vector of the user characteristics;
• applying the vector of user characteristics to the statistical model to identify an emotion to which the non-training user has the greatest probability of responding;
• and
• providing the non-training user with a promotional message that includes language associated with the identified emotion.

A computer system for providing a promotional message to a user based on a predicted emotional response, the system comprising:

• one or more processors;
• one or more memory units coupled to the one or more processors, wherein the one or more memory units store instructions that, when executed by the one or more processors, cause the system to perform the operations of:
• creating a statistical model that, for each of a set of emotions, predicts the probability that a user with certain characteristics will respond to the emotion, wherein creating the model comprises:
• receiving training data related to a plurality of training users, wherein, for each training user, the training data received includes user characteristics, a plurality of promotional messages provided to the training user, and, for each of the messages, an indication of whether the training user responded to the message;
• automatically tagging each of the promotional messages in the training data with one or more emotions to identify the emotions received by the training users;
• and
• building the statistical model using a statistical machine learning algorithm, wherein the model is trained using the user characteristics of the training users, the emotions provided to the training users, and the corresponding response data;
• using machine-learning-based modeling to automatically predict the emotion to which a non-training user is most likely to respond and to use the prediction to provide a promotional message to the non-training user by performing the following:
• receiving user characteristics for a non-training user;
• creating a vector of the user characteristics;
• applying the vector of user characteristics to the statistical model to identify an emotion to which the non-training user has the greatest probability of responding;
• and
• providing the non-training user with a promotional message that includes language associated with the identified emotion.

A non-transitory computer-readable medium comprising a computer program, that, when executed by a computer system, enables the computer system to perform the following method for providing a promotional message to a user based on a predicted emotional response, the method comprising:

• receiving user characteristics for a user;
• creating a vector of the user characteristics;
• accessing a statistical model that, for each of a set of emotions, predicts the probability that a user with certain characteristics will respond to the emotion;
• applying the vector of user characteristics to the statistical model to identify an emotion to which the non-training user has the greatest probability of responding;
• and
• providing the non-training user with a promotional message that includes language associated with the identified emotion.