Cognitive Assessment Using EEG Data: Developing a Brain-Computer İnterface for Cognitive Function Evaluation

Electroencephalography (EEG) is a non-invasive real-time technique for monitoring brain activity, offering insights into cognitive processes such as attention, memory, and problem-solving. A methodology for developing a brain-computer interface for cognitive function evaluation using EEG data is presented. The methodology involves data preprocessing, feature extraction, optimisation, machine learning model development, testing, and validation. Novel attention indexes based on the power spectral density ratios of theta, alpha, and beta frequency bands are proposed, and the efficiency of four machine learning algorithms in predicting attention spans from EEG features is evaluated. A 10-fold cross-validation shows a mean accuracy of 93.96%, indicating promising EEG features for predicting attention span.