A Dataset and Methodology for Self-Efficacy Feeling Prediction During Industry 4.0 VR Activity

Virtual Reality Learning Environments (VRLE) have advantages in training contexts. However, VRLE lacks of User-adaptive system which adapt scenario to the user's state. As there is a lack of multi-sensor dataset, this paper presents the IVRASED dataset collected in an industrial VRLE with the following sensors: electroencephalogram (EEG), eye-tracking (ET), galvanic skin response (GSR) and electrocardiogram (ECG). Classification of the user's state is performed with a deep learning architecture and the results show an accuracy of 77.8% for the best sensors combination.