Abstract
Several statistical and machine learning methods are proposed to estimate thetype and intensity of physical load and accumulated fatigue . They are based onthe statistical analysis of accumulated and moving window data subsets withconstruction of a kurtosis-skewness diagram. This approach was applied to thedata gathered by the wearable heart monitor for various types and levels ofphysical activities, and for people with various physical conditions. Thedifferent levels of physical activities, loads, and fitness can bedistinguished from the kurtosis-skewness diagram, and their evolution can bemonitored. Several metrics for estimation of the instant effect and accumulatedeffect (physical fatigue) of physical loads were proposed. The data and resultspresented allow to extend application of these methods for modeling andcharacterization of complex human activity patterns, for example, to estimatethe actual and accumulated physical load and fatigue, model the potentialdangerous development, and give cautions and advice in real time.