Sneaky emotions: impact of data partitions in affective computing experiments with brain-computer interfacing

• Published in: Biomedical engineering letters Vol. 14; no. 1; pp. 103 - 113
• Main Authors: Moreno-Alcayde, Yoelvis, Traver, V. Javier, Leiva, Luis A.
• Format: Journal Article
• Language: English
• Published: Korea The Korean Society of Medical and Biological Engineering 01.01.2024; Springer Nature B.V
• Subjects: Affective computing; Annotations; Biological and Medical Physics; Biomedical Engineering and Bioengineering; Biomedicine; Biophysics; Brain; Classification; Computer applications; Data points; Data transfer (computers); EEG; Electroencephalography; Emotion recognition; Emotions; Engineering; Feature extraction; Human-computer interface; Implants; Machine learning; Medical and Radiation Physics; Original; Original Article; Partitioning; Segments; Video; Emotion recognition; BCI; EEG; Data splits; Videos

Brain-Computer Interfacing (BCI) has shown promise in Machine Learning (ML) for emotion recognition. Unfortunately, how data are partitioned in training/test splits is often overlooked, which makes it difficult to attribute research findings to actual modeling improvements or to partitioning issues....