We present an autonomous artificial intelligent agent to dynamically communicate to a high performance athlete how much time remains for that athlete to sustain his current effort in watts, and then, if the result is incorrect, to autonomously recalibrate its algorithms to not make that same mistake again, it takes a series of experiments to ascertain what algorithms are most suited for this goal. The field of measuring elite athletes's abilities is based upon measurements such as maximal O2 consumption (V O2max), Heart beats per minute (bpm), Watts of energy (watts and kilojoules) and peak power output ((Ppeak)) to name a few. Because there is an element of uncertainty and imprecision in measuring human performance, this field of art suits the likes of Rough Set Theory. The aim of this paper is to demonstrate that Rough Set Theory is remarkably adept at yielding accurate predictions of performance limits at a particular time. The aforementioned leads one to the hypothesis that a Rough Set engine may, in the future, select the best Knowledge Discovery in Database (KDD) tool form a plurality of KDD, fuzzy, neural and machine learning tools - to eventually become intelligent in predicting the performance limits of high performance athletes. This paper first reviews lactate and blood issues in delivering O2 to muscles, then it presents the rough set hypothesis, experimentations and experiment results.
© Copyright 2017 2017 International Conference on Computational Science and Computational Intelligence (CSCI). Veröffentlicht von IEEE. Alle Rechte vorbehalten.