Leveraging Ensemble Techniques for Enhanced Eye State Classification with EEG Data

Brain-Computer Interaction (BCI) is an indispensable component in the constantly evolving field of human-machine interaction, offering direct interaction between the human brain and external devices. The acquisition of electroencephalogram, or EEG, signals by BCI devices is vital for this connection. The objective of this study is to differentiate eye states (open and closed) with the goal of shedding light on the consequences of artifacts when performing various BCI tasks. The background study explores the key concepts of BCI, emphasizing the critical role that EEG signals play in the intricate relationship between humans and computers. Within the BCI framework, the problem statement accentuates the importance of eye state classification across tasks. In an effort to improve classification accuracy and eliminate artifacts, this research strategically integrates techniques through the implementation of an extensive approach and assessments of multiple ensemble strategies. The findings and analysis reveal the efficiency of ensemble approaches in understanding eye states, hence improving BCI applications and expanding human-machine interaction.