Electroencephalography based User Mood Validation using Convolutional Neural Network

EEG signals are being used for medical diagnosis purposes since centuries. In the last few decades, EEG is also being used for user brainwave reading for various purposes in computer science and other related branches. In the same direction, attempts were made here to validate two classes of what the user is thinking with the help of CNN. The study was conducted and found that accuracy was having the scope of improvement while having the same task being implemented using ANN and other machine learning techniques such as SVM, Decision Tree, Random Forests, etc. After finding these techniques to be less effective in finding sufficient accuracy, CNN comes to help to find better accuracy. The research here surrounds binary classification with exceptional accuracy of 99.82% with a dataset of 14 participants.