EMG Signal RNN Classifier

By using a dataset of electromyography (EMG) signals from athletes performing an acrobatic move in four different positions, a predictive Machine Learning model was developed. This model is capable of classifying the position in which the trained acrobatic move was performed when receiving a new EMG signal.

The purpose of this project was to create my Computer Science’s thesis that determines the most effective Machine Learning tool for analyzing these signals (Neural Networks, XGBoost, Logistic Regression, Random Forest, and SVM) by measuring Accuracy, Cross-Validation Accuracy, and calculating the ROC curve for each model to quantify their performance.