Recognition of gait motion by using data mining

The purpose of this research is to embody feature parameters of Suriashi (sliding gait motion in Japanese traditional performing arts and Japanese traditional martial arts) by multivariate data analysis and to recognize Suriashi movement among other gait motions by using SVM (Support Vector Machine). Experiments of gait motion were carried out using motion capture on the Suriashi of Japanese Traditional Dance and Noh and ordinary human gait. We are able to analyzing the principal parameter to distinguish the Suriashi in various gait motions by conducting PCA (Principal Components Analysis) and cluster analysis of parameters of gait motion. It was found that gait motion of Suriashi can be recognized by the principal parameters to discriminate among the different gait motions by using SVM. In addition, it is expected that our research will help practitioners and masters of Japanese Traditional Dance and Noh with Suriashi training through giving new information of the principal parameters on Suriashi movements.