Attendee engagement determining system and method

A system, comprising:

• a plurality of computing devices including a plurality of attendee computing devices each having a display to register a plurality of particular attendees for a future event and a presenter computing device having a display to display an engagement level for each attendee of the future event;
• a backend computer system that receives a set of registration data from each attendee computing device when the particular attendee registers for the future event;
• the backend computer system having a processor, memory and a plurality of lines of computer code that are configured to:
• retrieve a feature vector for a particular attendee, the feature vector having a score for each of a plurality of features for the particular attendee, the feature vector being generated based on past event data of the particular attendee;
• generate an engagement prediction for the particular attendee for the future event using a classifier model, a decision tree classifier and a deep neural network and the feature vector for the particular attendee, wherein the backend is further configured to generate one of a high engagement output and a low engagement output for each of the decision tree classifier and the deep neural network, combine the outputs of the decision tree classifier and the deep neural network to determine the engagement prediction is one of a high engagement prediction output and a low engagement prediction output and execute a voting process to determine the engagement prediction in which the high engagement prediction output is selected if at least one output of the decision tree classifier and the deep neural network is the high engagement output and the low engagement prediction output is selected if both of the outputs of the decision tree classifier and the deep neural network are low;
• and
• display a user profile of the particular attendee on a display of the presenter computing device, the user profile including the engagement prediction of the particular attendee for the future event.

An apparatus, comprising:

• a backend computer system that receives a set of registration data from each attendee computing device when a particular attendee registers for a future event, the backend computer system having a processor, memory and a plurality of lines of computer code that are configured to:
• retrieve a feature vector for the particular attendee, the feature vector having a score for each of a plurality of features for the particular attendee, the feature vector being generated based on past event data of the particular attendee;
• generate an engagement prediction for the particular attendee for the future event using a classifier model, a decision tree classifier and a deep neural network and the feature vector for the particular attendee, wherein the computer system is further configured to generate one of a high engagement output and a low engagement output for each of the decision tree classifier and the deep neural network, combine the outputs of the decision tree classifier and the deep neural network to determine the engagement prediction is one of a high engagement prediction output and a low engagement prediction output and execute a voting process to determine the engagement prediction in which the high engagement prediction output is selected if at least one output of the decision tree classifier and the deep neural network is the high engagement output and the low engagement prediction output is selected if both of the outputs of the decision tree classifier and the deep neural network are low;
• and
• communicate a user profile of the particular attendee to a presenter, the user profile including the engagement prediction of the particular attendee for the future event.