SIMBAD references

Type IIP and Type IIL supernovae are defined on the basis of their light curves, but the spectral criteria for distinguishing these two types of supernovae (SNe) remain unclear. We propose a spectral classification method. First, we subtract the principal components of different wavelength bands in the spectra based on the functional principal components analysis method. Then, we use support vector machine and artificial neural network to classify these two types of SNe. The best F1_Score of our classifier is 0.871 for SNe IIL, and 0.974 for SNe IIP. We found that by only using the H_α line at 6150-6800 Å for classification, the F1_Score up to 0.961 for Type IIP, and 0.818 for Type IIL SNe can be obtained. These results indicate that the profile of the H_α spectral line is the key to distinguishing the two types of SNe.