Defining athlete's anaerobic and aerobic thresholds by using similarity measures and differential evolution

This paper presents a computational model for defining an athlete's aerobic and anaerobic thresholds. The model simulates the decision-making done by specialists in sport medicine. Computation is based on fuzzy decision making using expert criteria. Decision-making means in this paper the use of similarity measures, generalized mean, fuzzy membership functions and differential evolution. We have used differential evolution to tune parameters in our comparison measure based on similarities combined with generalized mean. The comparison measure is used to combine our models fuzzified criterions, which originally came from experts. Simulation results show that this model succeeds in finding thresholds, which does not differ statistically significantly from the thresholds estimated by human sport medicine experts.