Document Type



The purpose of this study was to determine the functional groups in different levels during stance phase of running. 118 students (58 males and 60 females) ran in two footwear conditions (Nike free5 and Vibram FiveFingers shoes) and barefoot. Mechanical energy of pelvic, thigh, leg and right foot were calculated. Functional groups were determined using principal component analysis, self-organizing maps, k-means clustering and support vector machine methods based on lower limb mechanical energy. Five first level functional groups were defined in barefoot, Nike and FiveFinger running conditions with accuracy of 95.80%, 91.60% and 91.60%, respectively. 41 subjects were identified as the third level functional groups. According to our results, the functional groups were well recognized with the use of dimension reduction and unsupervised clustering methods.