EEG Recordings with Eye State Classification

EEG Eye State data from the UCI Machine Learning Repository provides brainwave recordings linked to whether a subject's eyes were open or closed. The dataset is used for brain-computer interface research, though the exact number of samples and recording details are unspecified. It originates from the UCI repository, a common source for benchmark machine learning data.

Use Cases

• Classify eye state (open/closed) from 14 EEG channel time-series data using a support vector machine.
• Analyze temporal patterns in EEG signals preceding or following eye state transitions.
• Benchmark feature extraction methods for EEG signals against a binary classification target.
• Train a recurrent neural network to model the sequential dependencies in the EEG time-series for state prediction.

Strengths

• Dataset is hosted on the UCI repository, a trusted source for benchmark machine learning data.
• Includes a clear binary classification target (eye state) linked to multivariate time-series input.

Limitations

• Sample size, recording duration, and subject demographics are unknown, limiting reproducibility assessment.
• Lack of column descriptions prevents detailed understanding of specific EEG channels and preprocessing steps.

Provenance

Source

UCI Machine Learning Repository

Collection Method

Recorded via a Neuroscan brain-computer interface headset, according to UCI tags.