Eating Gesture Recognition

We used the PCA as new feature set by multiplying both and divided this new feature set into two parts for each user: Part 1: Training and Part 2: Test. We used 60% of the data for each user as training and the rest of 40% as test data. We used three types of machines in this assignment: a) Decision trees, b) Support vector machines and c) Neural networks. We trained each machine with the training data and then used the test data to report accuracy. Using the accuracy metrics of Precision, Recall, F1 score and ROC reported each metric for every group.