人体咀嚼规律研究.pdf

Analysis on Chewing Patterns Statistics 504 Chuqiao iu, Zhenyu Jia, Bo Zhang Motivation Given growing interest in functional data analysis (FDA) as a useful method for analyzing human movement data, it is critical to understand the effects of standard FDA procedures on biomechanical analyses. In our study, we would like to examine whether people’s chewing behaviors follow certain unknown patterns and whether FDA will make an impact on the clustering result. First of all, we transformed the given data into functional form. We looked at the curves by our own eyes and picked out the most recognizable features. We then applied clustering techniques and tried to interpret the result. Hopefully by clustering we can identify a few typical patterns that people’s chewing behaviors have, and we hope that these generic patterns will be useful for biomechanics in future studies, in order to explore more characteristics of chewing movements of human beings. We will also show in a short while by comparison that fitting the data using functional data analysis techniques gives us a better clustering result, than simply using the raw data or local smoothing. Data Data Collection Twenty-two adults (2 males, 20 females) between age 19 and 58 participated in the study. All of them had fully functional teeth, and had no other symptoms or illness that might harm the result of the experiment. There were in total 9 markers attached to different parts of heads used to track jaw, chin and head movements. The jaw data were obtained by the three markers attached to the lower left canine (K-nine). The