Classification of High Time Resolution Universe Survey 2 Data by Machine Learning Technique | Original Article

ABSTRACT:

Machine learning techniques are useful to finding or discovering the new pulsar. We study the candidate filters used to moderate theseproblems for the period of past few years. Pulsars are types of star of huge scientific interest. We have used two classification techniques like Conditional Inference Tree (CTREE) and Classification and Regression Tree (CART) and their ensemble model (CTREE+CART) for classifying the HTRU2 dataset. The ensemble model (CTREE+CART) give the better performance compared to the Individual models of each classifiers. The ensemble model (CTREE+CART) is useful of the candidates must be classified in to pulsar and non-pulsar classes to aid discovery.