Hybrid Attention-Infused Ensemble Learning Model for EEG-Based Workload Assessment

Accurate and real-time assessment of cognitive workload is vital for enhancing human performance and safety in complex task environments such as aviation, driving, and human-computer interaction. Electroencephalogram (EEG) signals, due to their high temporal resolution and direct link to brain activity, have emerged as a promising modality for non-invasive workload monitoring. However, the non-stationary, nonlinear, and multi-dimensional nature of EEG signals poses significant challenges for conventional machine learning approaches. Therefore, this paper proposes the analysis of EEG-based workload assessment using a hybrid attention-infused ensemble learning model. An EEG dataset is first pre-processed using a Finite Impulse Response (FIR) filter to remove the high-frequency noises. The feature extraction using Hilbert-Huang Transform (HHT) is used to extract the relevant features and patterns. After that, the classification method of the attention-infused ensemble learning model is employed to detect the EEG-based workload assessment. This classification integrates attention mechanisms and an infused ensemble of classifiers for enhancing feature relevance, accuracy, and interpretability. Additionally, the classification algorithm has achieved high performance and accuracy. With a classification accuracy of 97% and a precision of 99%, are demonstrated by simulations using Python software. This system framework provides strong potential for real-world applications in adaptive interfaces, fatigue monitoring, and neuroergonomic systems.