The ICLabel project introduces an automated electroencephalographic (EEG) independent component classifier, a dataset, and a website. The classifier, available for MATLAB, improves upon existing methods by enhancing accuracy and computational efficiency. The dataset contains over 200,000 spatiotemporal measures for ICs from more than 6,000 EEG recordings, along with matching component labels for over 6,000 ICs. The website facilitates crowdsourcing of IC labels and serves as an educational tool. The classifier outperforms or matches previous methods in terms of accuracy while being significantly faster, achieving ten times the speed of the best previous method. The project addresses the challenges of noise sensitivity and ambiguity in ICA results, providing a comprehensive solution for EEG data analysis.The ICLabel project introduces an automated electroencephalographic (EEG) independent component classifier, a dataset, and a website. The classifier, available for MATLAB, improves upon existing methods by enhancing accuracy and computational efficiency. The dataset contains over 200,000 spatiotemporal measures for ICs from more than 6,000 EEG recordings, along with matching component labels for over 6,000 ICs. The website facilitates crowdsourcing of IC labels and serves as an educational tool. The classifier outperforms or matches previous methods in terms of accuracy while being significantly faster, achieving ten times the speed of the best previous method. The project addresses the challenges of noise sensitivity and ambiguity in ICA results, providing a comprehensive solution for EEG data analysis.