Table 2 The best prediction results from classification models with all reconstruction algorithms
From: An advanced scheme of compressed sensing of acceleration data for telemonintoring of human gait
Reconstruction algorithms | Classification models | Prediction results | ||||
|---|---|---|---|---|---|---|
ACC (%) | Upstairs | Downstairs | ||||
Sen (%) | Spe (%) | Sen (%) | Spe (%) | |||
OMP | KStar | 62.5 | 60.9 | 64.7 | 64.7 | 60.9 |
SVM | 65 | 60 | 66.7 | 66.7 | 60 | |
MLP | 67.5 | 65.2 | 70.6 | 70.6 | 65.2 | |
BP | KStar | 67.5 | 65.2 | 70.6 | 70.6 | 65.2 |
SVM | 70 | 65.4 | 78.6 | 78.6 | 65.4 | |
MLP | 72.5 | 73.7 | 71.4 | 71.4 | 74.7 | |
SP | KStar | 55 | 87.5 | 57.1 | 57.1 | 87.5 |
SVM | 62.5 | 60.9 | 64.7 | 64.7 | 60.9 | |
MLP | 65 | 62.5 | 68.8 | 68.8 | 62.5 | |
SL0 | KStar | 70 | 65.4 | 78.6 | 78.6 | 65.4 |
SVM | 72.5 | 69.6 | 76.5 | 76.5 | 69.6 | |
MLP | 75 | 67.9 | 91.7 | 91.7 | 67.9 | |
DGS | KStar | 67.5 | 66.7 | 68.4 | 68.4 | 66.7 |
SVM | 72.5 | 69.6 | 76.5 | 76.5 | 69.6 | |
MLP | 75 | 75 | 75 | 75 | 75 | |
Group Lasso | KStar | 72.5 | 69.6 | 76.5 | 76.5 | 69.6 |
SVM | 75 | 72.7 | 77.9 | 77.9 | 72.7 | |
MLP | 77.5 | 72 | 86.7 | 86.7 | 72 | |
SOMP | KStar | 77.5 | 76.5 | 79 | 79 | 76.2 |
SVM | 85 | 79.2 | 93.8 | 93.8 | 79.2 | |
MLP | 87.5 | 85.7 | 89.5 | 89.5 | 85.7 | |
BSBL-BO | KStar | 80 | 82.6 | 94.1 | 94.1 | 82.6 |
SVM | 92.5 | 90.5 | 94.7 | 94.7 | 90.5 | |
MLP | 95 | 95 | 95 | 95 | 95 | |