ABSTRACT

This study developed and evaluated machine learning algorithms to predict children’s physical activity category from raw accelerometer data collected at the hip. Fifty participants (mean age = 13.9 ± 3.0 y) completed 12 activity trials that were categorized into 5 categories: sedentary (SED), light household activities and games (LHHAG), moderate-vigorous games and sports (MVGS), walking (WALK), and running (RUN). Random Forest (RF) and Logistic Regression (LR) classifiers were trained with features extracted from the vector magnitude using 10 s non-overlapping windows. Classification accuracy was evaluated using leave-one-subject-out cross validation. Overall accuracy for the RF and LR classifiers was 95.7% and 94.3%, respectively. Classification accuracy was excellent for SED (96.3% – 98.1%), LHHAG (92.3% – 95.2%), WALK (94.5% – 97.1%), RUN (99.5% – 99.6%); and MVGS (87.5% – 92.7%). The results indicate that classifiers trained on features in the raw acceleration from the hip can be used for activity recognition in young people.

Abbreviations: VM: Vector Magnitude; RF: Random Forest; LR: Logistic Regression; LOSO: Leave-One-Subject-Out

摘要

这项研究开发并评估了机器学习算法，可以从在臀部收集的原始加速度计数据预测儿童的体育活动类别。50名参与者（平均年龄= 13.9±3.0岁）完成了12项活动试验，分为5类：久坐（SED），轻度家庭活动和游戏（LHHAG），中度运动和运动（MVGS），步行（WALK） ，然后运行（RUN）。随机森林（RF）和逻辑回归（LR）分类器使用10 s非重叠窗口从矢量幅度中提取的特征进行训练。分类准确度使用留一法制交叉验证进行评估。RF和LR分类器的总体准确度分别为95.7％和94.3％。SED（96.3％– 98.1％），LHHAG（92.3％– 95.2％），WALK（94.5％– 97.1％），RUN（99）的分类准确度极佳。5％– 99.6％）；MVGS（87.5％– 92.7％）。结果表明，针对髋关节原始加速度特征进行训练的分类器可用于年轻人的活动识别。

缩写：VM：矢量幅度；RF：随机森林；LR：逻辑回归；漏写：离开一个主题