In this work, we propose the use of Kernel Principal Component Analysis (KPCA) combined with k = 1 nearest neighbor algorithm (1NN) as a framework for supernovae (SNe) photometric classification. It is specially recommended for analysis where the user is interested in high purity in the final SNe Ia sample. Our method provide good purity results in all data sample analyzed, when SNR≥5. As a consequence, we can state that if a sample as the Supernova Photometric Classification Challenge were available today, we would be able to classify ≈ 15% of the initial data set with purity higher than 90%. This makes our algorithm ideal for a first approach to an unlabeled data set or to be used as a complement in increasing the training sample for other algorithms. Results are sensitive to the information contained in each light curve, as a consequence, higher quality data (low noise) leads to higher successful classification rates.
展开▼
