Abstract: In some embodiments, a method includes receiving longitudinal interaction data. The longitudinal interaction data is sorted into a plurality of bins. Using a run-length encoder, at least one inactivity period indication is generated, and an inferred period indication is generated based on the at least one inactivity period indication. The method also includes (1) generating, based on the plurality of bins, a plurality of activity metrics and (2) sorting, using a clustering model and based on the plurality of activity metrics, the plurality of bins to produce an inactivity cluster. An expected period indication and a characteristic value are generated based on (1) the inferred period indication and (2) the expected period indication. A series of actions is determined by providing, via the processor, the characteristic value as input to a machine learning model.

Abstract: A method includes receiving, at a processor, past state data and past activity data and generating (1) a covariance matrix based on a state model, an activity dissipation model, and an activity magnitude model, associated with a group including the user, (2) at least one activity impact dissipation parameter associated with the user, and (3) at least one activity magnitude parameter associated with the user. The method includes generating at least one state impact value based on the at least one activity impact dissipation parameter and the at least one activity magnitude parameter. The method includes generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value, calculating future state data based on the non-linear regression model, and transmitting at least one signal to cause display of the future state data.