WLnet: Towards an Approach for Robust Workload Estimation Based on Shallow Neural Networks

Electroencephalography (EEG) is a non-invasive technology used for the human brain-computer interface. One of its important applications evaluation mental state an individual, such as workload estimation. In previous works, common spatial pattern feature extraction methods have been proposed EEG-based detection. Recently, several novel were introduced to detect EEG workloads. However, it still unknown which one these that offers best performance this article, four extract features: (a) extraction; (b) temporally constrained sparse group (c) EEGnet; and (d) new shallow convolutional neural network estimation (WLnet). The classification accuracy was compared. Experimental results demonstrate WLnet achieved detection in both stress non-stress conditions. We believe may be relevant real-life