Generates simulated GC-MS datasets over short retention time intervals with varying rank (3, 4, 5 components) and noise levels (0.001, 0.01, 0.1). Pure components use mass spectra from real GC-MS data, Gaussian elution profiles with random retention time shifts up to 7 points, and random uniform concentrations. Produces 900 total datasets for benchmarking PARAFAC2 and other decomposition methods.
Tian, K.; Wu, L.; Min, S.; Bro, R. (2018). Geometric search: A new approach for fitting PARAFAC2 models on GC-MS data. Talanta, 185, 378-386. https://doi.org/10.1016/j.talanta.2018.03.088
GitHub repository. https://github.com/ucph-chemometrics/Simulated-GC-MS Version 1.0.