Description: Electrocardiography (ECG) is a transthoracic interpretation of the electrical activity of the heart over a period of time, which can reflect the physical condition of a heart. We carry a series of experiments based on ten sets of different ECG data. First, we pre-process the raw data with data processing technology, and then we use the feature extraction technology of PCA and ICA to get a new feature space. At last we use support vector machine to train a model which can and finally predict and classify the unlabeled ECG data. We validate and optimize the model by evaluating its performance when training it.
To Search:
File list (Check if you may need any files):
get_test_data.m
ica_classification.m
ica_predict_on_test.m
pca_classification.m
predict_on_test.m
train_test_data.m