Unsupervised star, galaxy, qso classification: Application of HDBSCAN

Classification will be an important first step for upcoming surveys that will detect billions of new sources such as LSST and Euclid, as well as DESI, 4MOST and MOONS. The application of traditional methods of model fitting and colour-colour selections will face significant computational constraints, while machine-learning (ML) methods offer a viable approach to tackle datasets of that volume. While supervised learning methods can perform very well for classification tasks, the creation of representative and accurate training sets is a resource and time consuming task. We present a viable alternative using an unsupervised ML method to separate stars, galaxies and QSOs using photometric data. The heart of our work uses HDBSCAN to find the star, galaxy and QSO clusters in a multidimensional colour space. We optimized the hyperparameters and input attributes of three separate HDBSCAN runs, each to select a particular object class, and thus treat the output of each separate run as a binary classifier. We subsequently consolidate the output to give our final classifications, optimized on their F1 scores. We explore the use of Random Forest and PCA as part of the pre-processing stage for feature selection and dimensionality reduction. Using our dataset of 50,000 spectroscopically labelled objects we obtain an F1 score of 98.9, 98.9 and 93.13 respectively for star, galaxy and QSO selection using our unsupervised learning method. We find that careful attribute selection is a vital part of accurate classification with HDBSCAN. We applied our classification to a subset of the SDSS spectroscopic catalogue and demonstrate the potential of our approach in correcting misclassified spectra useful for DESI and 4MOST. Finally, we create a multiwavelength catalogue of 2.7 million sources using the KiDS, VIKING and ALLWISE surveys and publish corresponding classifications and photometric redshifts.