MODELLING STABILOMETRIC TIME SERIES

Stabilometry is a branch of medicine that studies balance-related human functions. Stabilometric systems generate time series. The analysis of these time series using data mining techniques can be very useful for domain experts. In the field of stabilometry, as in many other domains, the key nuggets of information in a time series are concentrated within definite time periods known as events. In this paper, we propose a technique for creating reference models for stabilometric time series based on event analysis. After testing the technique on time series recorded by top-competition sportspeople, we conclude that stabilometric models can be used to classify individuals by their balance-related abilities.