System and a method for monitoring progress of a learner through an experiential learning cycle

A computer implemented method of monitoring progress of a learner through an experiential learning cycle comprising a plurality of learning phases, the method comprising:

• receiving, with a computer server accessible through a communications network, event data through the communications network from user computing devices, the received event data being generated by the learner progressing through the experiential learning cycle and a collaborating learner progressing through another experiential learning cycle, the experiential learning cycle and the another experiential learning cycle are interdependent cycles, the received event data being representative of data associated with activities of the learner in the experiential learning cycle;
• training, with the computer server, a multiclass decision forest model by feeding a selection of time windows of combination of training event data and empirical data indicative of the experiential learning cycle into a decision forest, wherein the training event data comprises historical event data generated by the learner and the collaborating learner, the empirical data comprises probabilistic vectors indicative of an influence of the collaborating learner being in a specific learning phase of the other experiential learning cycle on probabilities that the learner is in each of the plurality of learning phases of the experiential learning cycle, and the combination of the training event data and the empirical data are labelled with learning phases;
• executing, with the computer server, the trained multiclass decision forest model with the received event data as input to predict phase probabilities that the learner currently is in each of the plurality of learning phases of the experiential learning cycle, the phase probability being indicative of a likelihood that the learner is currently in an associated learning phase;
• determining, with the computer server, the associated learning phase with the highest phase probability among the plurality of learning phases as a learning phase of the experiential learning cycle in which the learner currently is;
• evaluating, with the computer server, the progress of the learner through the experiential learning cycle by comparing the determined learning phase with the empirical data;
• and
• in response to a result of the comparison meeting predetermined criteria, automatically triggering an intervention by generating information in relation to the intervention and communicating the information to related user computing devices.

A non-transitory machine-readable media, having instructions stored on the machine-readable media, the instructions configured to, when executed, cause a machine to:

• receive event data through a communications network from user computing devices, the received event data being generated by a learner progressing through a experiential learning cycle and a collaborating learner progressing through another experiential learning cycle, the experiential learning cycle comprising a plurality of learning phases, the experiential learning cycle and the another experiential learning cycle are interdependent cycles, the received event data being representative of data associated with activities of the learner in the experiential learning cycle;
• train a multiclass decision forest model by feeding a selection of time windows of combination of training event data and empirical data indicative of the experiential learning cycle into a decision forest, wherein the training event data comprises historical event data generated by the learner and the collaborating learner, the empirical data comprises probabilistic vectors indicative of an influence of the collaborating learner being in a specific learning phase of the other experiential learning cycle on probabilities that the learner is in each of the plurality of learning phases of the experiential learning cycle, and the combination of the training event data and the empirical data are labelled with learning phases;
• execute the trained multiclass decision forest model with the received event data as input to predict phase probabilities that the learner currently is in each of the plurality of learning phases of the experiential learning cycle, the phase probability being indicative of a likelihood that the learner is currently in an associated learning phase;
• determine the associated learning phase with the highest phase probability among the plurality of learning phases as a learning phase of the experiential learning cycle in which the learner currently is;
• evaluate the progress of the learner through the experiential learning cycle by comparing the determined learning phase with the empirical data;
• and
• in response to a result of the comparison meeting predetermined criteria, automatically trigger an intervention by generating information in relation to the intervention and communicating the information to related user computing devices.