Skip to main content
ARS Home » Plains Area » Houston, Texas » Children's Nutrition Research Center » Research » Publications at this Location » Publication #307563

Title: Support vector machines classifiers of physical activities in preschoolers

Author
item ZHAO, WEI - Drexel University
item ADOLPH, ANNE - Children'S Nutrition Research Center (CNRC)
item PUYAU, MAURICE - Children'S Nutrition Research Center (CNRC)
item VOHRA, FIROZ - Children'S Nutrition Research Center (CNRC)
item BUTTE, NANCY - Children'S Nutrition Research Center (CNRC)
item ZAKERI, ISSA - Drexel University

Submitted to: Physiological Reports
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/15/2013
Publication Date: 6/1/2013
Citation: Zhao, W., Adolph, A.L., Puyau, M.R., Vohra, F.A., Butte, N.F., Zakeri, I.F. 2013. Support vector machines classifiers of physical activities in preschoolers. Physiological Reports. 1(1):1-7.

Interpretive Summary: We wanted to develop, test, and compare two statistical methods (multinomial logistic regression (MLR) and support vector machines (SVM)) for classifying the kinds of physical activities typical of preschool-aged children based on data from activity monitors. We enrolled 69 children, ages 3-5, to wear an activity monitor and do some supervised physical activities. Data output from the activity monitor include accelerometer counts, steps, and child's position (standing, sitting, etc.). We developed 58 models based on the combinations of the output variables. We found that SVM of accelerometer data can be used to correctly classify physical activities in preschool children with an acceptable error rate. These finding support a more beneficial statistical method for this research.

Technical Abstract: The goal of this study is to develop, test, and compare multinomial logistic regression (MLR) and support vector machines (SVM) in classifying preschool-aged children physical activity data acquired from an accelerometer. In this study, 69 children aged 3-5 years old were asked to participate in a supervised protocol of physical activities while wearing a triaxial accelerometer. Accelerometer counts, steps, and position were obtained from the device. We applied K-means clustering to determine the number of natural groupings presented by the data. We used MLR and SVM to classify the six activity types. Using direct observation as the criterion method, the 10-fold cross-validation (CV) error rate was used to compare MLR and SVM classifiers, with and without sleep. Altogether, 58 classification models based on combinations of the accelerometer output variables were developed. In general, the SVM classifiers have a smaller 10-fold CV error rate than their MLR counterparts. Including sleep, a SVM classifier provided the best performance with a 10-fold CV error rate of 24.70%. Without sleep, a SVM classifier-based triaxial accelerometer counts, vector magnitude, steps, position, and 1- and 2-min lag and lead values achieved a 10-fold CV error rate of 20.16% and an overall classification error rate of 15.56%. SVM supersedes the classical classifier MLR in categorizing physical activities in preschool-aged children. Using accelerometer data, SVM can be used to correctly classify physical activities typical of preschool-aged children with an acceptable classification error rate.