Enhanced Cognitive Load Detection Through Cissa and Binary Harris Hawks Optimization of EEG Features

Algorithms for the identification of cognitive loads are essential for the improvement of the people’s interaction with computers and decide on many aspects of their life, in such spheres as education, healthcare, and cognitive science. The proposed technique of this research work is to advance the current technique of cognitive load detection via EEG data with the integration of CiSSA and BHHO. CiSSA is used to detect the desired frequencies in the EEG waves and excerpt the topographies from them with higher quality than when using the original signals. Binary Harris Hawks Optimization is used for the feature selection such that the most important features of the EEGs are used in the classification process. The proposed method is then applied on the two publicly available datasets of EEG data and overall, the accuracy in detecting the amount of cognitive load is better than the conventional techniques that have been adopted earlier. As evidenced from the results the CiSSA-BHHO framework improves both the feature extraction and selection making it a reliable tool in real time cognitive load assessment.