| CPC G06Q 10/1095 (2013.01) [G06N 3/044 (2023.01)] | 20 Claims |
|
1. A system for applying an artificial intelligence engine in real-time, the system comprising:
a facilitator device accessible by a human facilitator, the facilitator device comprising:
a memory device having computer-readable program code;
a communication device;
a processing device operatively coupled to the memory device and to the communication device, wherein the processing device is configured to execute the computer-readable code to:
receive a plurality of recommended parameters from a computing system;
receive inputs comprising a plurality of parameters of a meeting including one or more of: a meeting start time, a meeting location, a meeting duration, a meeting topic, and a list of teammate participant names;
convey the plurality of recommended parameters; and
facilitate the meeting among the facilitator device and one or more teammate participant devices;
at least one teammate participant device accessible by the teammate participants comprising:
a memory device having computer-readable program code;
a communication device;
a processing device operatively coupled to the memory device and to the communication device, wherein the processing device is configured to execute the computer-readable code to:
communicate with the facilitator device to facilitate the meeting; and
receive one or more inputs of a binary meeting score indicating that the meeting was either productive or not productive;
a network interconnecting the teammate participant device, the facilitator device, and a computing system operating the artificial intelligence engine in the form of a recurrent neural network (RNN), the computing system comprising:
a memory device having computer-readable program code;
a communication device;
a processing device operatively coupled to the memory device and the communication device, wherein the processing device is configured to execute the computer-readable code to:
train, using a first set of training test data, the RNN to recommend the plurality of recommended parameters based on a prediction of an optimal meeting outcome, the training using an iterative training and testing loop that iteratively tests the first set of training test data compared to a target variable and makes adjustments in subsequent iterations to improve predictability of the target variable thereby improving accuracy of the RNN:
deploy the trained RNN;
transmit, via the network, the plurality of recommended parameters recommended by the RNN to the facilitator device;
receive from the facilitator device the plurality of parameters of the meeting;
receive from the at least one teammate participant device the binary meeting score;
correlate the binary meeting score to each of the plurality of parameters of the meeting to create parameter scores;
store the parameter scores in the memory device and provide the parameter scores in a feedback loop as a second set of training test data;
retrain the RNN using the second set of training test data:
deploy the retrained RNN:
receive from the facilitator device an additional plurality of parameters of a new meeting;
transmit, via the network, new recommended parameters for the new meeting:
receive from the teammate participant device a binary additional meeting score;
correlate the binary additional meeting score with the additional plurality of parameters of the new meeting to create additional parameter scores;
store the additional parameter scores in the memory device; and
provide the additional parameter scores to the feedback loop for additional training of the RNN.
|