© 2018 ISSN. This paper discusses the methodology for extracting and classifying a style and characteristic component from a walking motion. The walking motion is measured using four wearable motion sensors for acquiring segmented body motion. To extract the style and characteristic component, we use the singular value decomposition of the measured data and evaluate the contribution of each sensor module for gait identification by using the degree of class separation. From these results, the characteristic component of human gait features can be extracted by using singular vectors of whole data of walking motion. In addition, the singular vectors of higher order modes can be used for identifying individuals by proper choice of the modes. Furthermore, using the degree of class separation, the important body segments for gait identification can be indicated by combinations of sensors with the high degree of separation.