DEEP LEARNING FOR FEATURE CLASSIFICATION OF EEG TO ACCESS STUDENT’S MENTAL STATUS

Electroencephalography (EEG) analysis has been a significant tool in neuroscience with uses of, neural designing and Brain–PC interfaces. A large number of the explanatory tools are utilized in EEG contemplates have utilized machine learning to reveal significant data for neural classification and neuro imaging. As of late, the accessibility of enormous EEG informational indexes and advances in machine learning have both prompted the arrangement of deep learning structures, particularly in the examination of EEG signals and the usefulness of cerebrum. In our proposed work, cuttlefish optimized (CFO) deep neural networks are utilized to distinguish depression in individuals by analyzing EEG signals. Best highlights of alpha, beta, theta values chose by CFO so as to improve an accuracy. The execution is assessed utilizing the Database for Emotion Analysis Physiological Signals (DEAP), which is an open EEG dataset. A list of capabilities is removed in 32 EEG channels, which comprises of measurable highlights, Hjorth parameters, band power, and frontal alpha asymmetry. MATLAB tool has been utilized to assess the execution of proposed framework.