Application of Machine Learning Approaches for Classifying Sitting Posture Based on Force and Acceleration Sensors

Figure 5

Classification accuracy for different explanatory variable combinations (scenarios A–F) using RF (NVarToSample = 4): columns 1–16 represented the pressure sensors (Supplementary Materials: Figure A1) and column 17 represented the backrest angle. The sensors on the seat pan are presented in a plain colour, those on the armrests are presented as crosshatched ones, those on the backrest are presented as normal hatched ones, and the backrest angle is in patterned dots.