Skip to main content
. 2019 Feb 6;7(2):e11270. doi: 10.2196/11270

Figure 12.

Figure 12

Importance of features in terms of the mean decrease in the Gini index in each classification task and in terms of the increase in node purity for the metabolic equivalent scores estimation regression task, as reported by the random forest model and 16-second window length for feature extraction (note the different range for the x-axis). METs: metabolic equivalent scores; IncNodePurity: Increase in node purity; cor_x_z: correlation between x-axis and z-axis; cor_y_z: correlation between y-axis and z-axis; cor_x_y: correlation between x-axis and y-axis; cv: covariance of the vector magnitude; sd_a: SD of angle; mean_a: mean of angle; pow.625: percentage of the power of the vector magnitude that is in the range of 0.6-2.5 Hz; dom.freq: dominant frequency of vector magnitude.