Raw EEG data for: Unsupervised learning for brain-computer interfaces based on event-related potentials: Review and online comparison

When you use the data of this repository please cite the following article:

Hübner, David, et al. "Unsupervised learning for brain-computer interfaces based on event-related potentials: Review and online comparison [research frontier]." IEEE Computational Intelligence Magazine 13.2 (2018): 66-77.

This dataset is similar to the one described in https://doi.org/10.5281/zenodo.192684

The difference is, in this experiment N=12 subjects had to use a visual speller to spell a 35-letter sentence followed by 35-letter free spelling, i.e., Run1-5 were copy-spell and Run6-10 was free spelling. Each subject did this three times (Block1-3) where the online used unsupervised classifier was reset at the start of each run.

There is a GitHub repository (TODO LINK) available if you want to use this dataset in MOABB (a framework to benchmark classifiers typically used in brain-computer interfaces). Additionally, an implementation of the learning from label proportions approach is available as well, in order to reproduce the online setup.