2019/07/19 by Daniel Lopez-Martinez, Lopez-Martinez, Daniel, Neska Elhaouij +3 · 1 citation
Neuroscience · Psychology · #EEG and Brain-Computer Interfaces #Emotion and Mood Recognition #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Sleep and Work-Related Fatigue
paper · pdf · doi:10.48550/arxiv.1907.09929
openalex publication_date 2019/07/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Affective states have a critical role in driving performance and safety. They\ncan degrade driver situation awareness and negatively impact cognitive\nprocesses, severely diminishing road safety. Therefore, detecting and assessing\ndrivers' affective states is crucial in order to help improve the driving\nexperience, and increase safety, comfort and well-being. Recent advances in\naffective computing have enabled the detection of such states. This may lead to\nempathic automotive user interfaces that account for the driver's emotional\nstate and influence the driver in order to improve safety. In this work, we\npropose a multiview multi-task machine learning method for the detection of\ndriver's affective states using physiological signals. The proposed approach is\nable to account for inter-drive variability in physiological responses while\nenabling interpretability of the learned models, a factor that is especially\nimportant in systems deployed in the real world. We evaluate the models on\nthree different datasets containing real-world driving experiences. Our results\nindicate that accounting for drive-specific differences significantly improves\nmodel performance.\n