Skip to main content

Table 1 Comparison between classification methods for eight classes of activities

From: Real-time motion onset recognition for robot-assisted gait rehabilitation

Classifier Prediction speed (obsa/s) F1-score (%) (LOPOCV)
Decision tree 4,006,000 \(78.87 \pm 11.6\)
KNN (neighbors = 3) 10,060 \(78.34\pm 8.4\)
Linear SVM 5,120,000 \(85.40 \pm 5.07\)
Linear logistic regression 4,067,000 \(83.9\pm 6.4\)
Neural Network 2,268,000 \(86.83\pm 6.2\)
  1. aObservation