Finding Accuracies of Various Machine Learning Algorithms by Classification of Pulsar Stars

In order to rank supervised machine learning techniques according to their accuracy, a number of them were applied on the HTRU2 dataset. Pulsars are exotic neutron stars rotating at very high RPMs which lead to a high scientific interest into recognising actual pulsars from a pool of candidates. False positives are almost indistinguishable from real positives and are generated most often due to internal and external noise and interference factors. The use of aforementioned ML techniques helps mitigate some of those problems. The raw observational data was collected by the High Time Resolution Universe Collaboration using the Parkes Observatory, funded by the Commonwealth of Australia and managed by the CSIRO. Deep investigations into the nature of exotic stars seem imminent, and a ranked list of the most accurate ML techniques presented in this paper will no doubt benefit the field of pulsar astronomy. A direct combination of future observatories and ML computation might yield unexpected results, and that is what we expect from this paper. We hope our work contributes in enabling scientific discoveries as humanity is finally becoming more and more capable of turning their heads up and understanding the mysteries of what lies beyond home.