STEM Progress Report-Ryan Wang

I was reading the research papers you provided and they gave insight for EEG application. Applications like music configuration based on emotional state and EEG based control over robots are applications that stood out because of their data classification methods. Although I did not fully comprehend their methods, I noticed that the group researching music modification based on emotional state created parameters using valence and arousal as x and y axis on a graph, respectively. The quadrants created by these axes are plotted accordingly so that happy, angry, sad, and relaxed lie on quadrants one, two, three, and four; respectively. With their classification methodology in mind, we decided to pursue a classification method that isolated alpha and beta waves, specifically their average amongst an epoch of 0.5s, plotted along a Cartesian graph so that the data (amplitude avg.)  is translatable into coordinate values (alpha, beta). Using the classier training tool, we hope to find the correct class labels in conjunction with linear discriminant analysis (LDA). With the class labels set correctly, we will proceed with LDA, which utilizes statistics to find a linear equation that is able distinguish if the controlled action is performed, according to neural activity, or not.  We plan to design an experiment to implement said classification immediately after an accurate application to classify data is established.