This notebook was made as the final project for my Neural Data Science course where I worked with team members to classify and predict epileptic seizures from EEG data. Due to the large amount of available literature, we thought it would be a good exercise to see if we could reach similar accuracy numbers to published papers (which we did!).
We contrasted a variety of dimensionality reduction methods and re-implemented many of the commonly used pipelines in existing EEG seizure detection research in an ablation study. Then, we used the conclusions of the ablation study to decide on a t-SNE -> Random Forest pipeline which ended up being our most accurate classification method.