|
|||||
|
|
Analysing expertise through data mining: an example based on treading water |
|
| Authors: | Christophe Schnitzler Chris Button Ludovic Seifert James Croft |
| Institution: | 1. LISEC, EA 2310, ESPé, Université de Strasbourg, Strasbourg, Francecschnitzler@unistra.fr;3. School of Physical Education, Sport and Exercise Sciences, University of Otago, Dunedin, New Zealand;4. CETAPS UPRES EA 3832, Faculty of Sports Sciences, University of Rouen, Mont Saint Aignan, France |
| Abstract: | AbstractA classification system of treading water based on a conceptual typology was first established and then verified empirically. The typology was established on two concepts: the nature of the forces created within the water and the type of inter-limb coordination used. Thirty-eight participants were videotaped while treading water. Multivariate statistics were used to understand how the different behavioural types related to expertise. Three distinct groups of coordination patterns were adopted during treading water. A support vector machine procedure was used as a confirmatory procedure. The data mining process provides a methodological framework to analyse expertise in sports activities, and in this context suggests that a taxonomy can be established among the numerous coordination solutions that allow humans to create stabilising forces in the water. |
| Keywords: | data mining machine learning typology constraint-led approach |