Machine Learning Based Hand Gesture Recognition via EMG Data

Electromyography (EMG) data gives information about the electrical activity related to muscles. EMG obtained from arm through sensors helps understand hand gestures. For this work, gesture were taken UCI2019 dataset MYO thalmic armband classied with six dierent machine learning algorithms. Articial Neural Network (ANN), Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), Naive Bayes (NB), Decision Tree (DT) and Random Forest (RF) methods preferred for comparison based on several performance metrics which are accuracy, precision, sensitivity, specicity, classication error, kappa, root mean squared error (RMSE) correlation. The belongs seven 700 samples 7 classes (100 per group) used in experiments. splitting ratio was 0.8-0.2, i.e. 80% of training 20% testing phase classier. NB found be best among other because high accuracy (96.43%) sensitivity lowest RMSE (0.189). Considering results parameters, it can said that study recognizes classies gestures successfully literature.