Comparative Analysis of Machine Learning Algorithms for EMG Signal Classification

We present a comparative analysis of different machine learning algorithms for classifying electromyography (EMG) signals for the application of prosthetic arm development. We applied four machine learning algorithms, namely Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Decision Trees, and Logistic Regression. These algorithms classified the signals into three classes mentioned. Our results show that SVM achieved the highest accuracy at 83 %, followed by KNN (77 %), Logistic Regression (75 %), and Decision Tree (74 %). These findings suggest that SVM is the most promising approach for EMG signal multiclass classification for the mentioned application. While these results provide a solid foundation, we discuss potential improvements through hyperparameter tuning and study with real EMG sensor data for prosthetics movement. This study contributes to the development of more responsive and intuitive machine learning-powered prosthetic arms, potentially enhancing the quality of life for individuals with upper limb amputations.